Skip to main content

Full text of "The Practice Of Lubrication"

See other formats


B]<OU 168498 >m 


CallNa 6'6& l/f** Accession^. 
Auther t?*>"*~&~ T- 

This book should be returned on or bdfore ihe datelast markedl)eloiv. 








T. C. THOMSEN, B.Sc. (Copenhagen) M.I.Mech.E., etc. 

Managing birector, T. Kriiger, Ltd., Copenhagen, Denmark; Independent 
Consulting Lubricating Engineer. For Many Years Chief Engineer to the 
Vacuum Oil Company, Ltd., London; Late Research Engineer 
to the Anglo-Mexican Petroleum Co., Ltd. 




COPYRIGHT, 1920, 1926, 1937, BY THE 


All rights reserved. This book, or 

parts thereof, may not be reproduced 

in any form without permission of 

the publishers. 









The development which has taken place in all directions since 
the previous edition of this book has necessitated considerable 
alterations and the addition of new matter. 

In producing lubricating oils, the oil refiner is less dependent 
than hitherto on the nature of the crude oil. .By means of newer 
and improved refining methods, he can eliminate undesirable 
constituents in the oil more efficiently than was formerly the case. 

It has thus become possible to obtain with greater ease, and 
from a greater variety of crudes, lubricating oils with non- 
carbonizing and nonemulsifying properties. 

Of new ideas which are developing may be mentioned extreme 
pressure lubricants and the addition of anticorrosion inhibitors 
to automobile oils. Colloidal graphite is also finding more 
extensive use. 

The development of lubricating oils appears to be in the direc- 
tion of producing inert lubricating media of suitable viscosities, 
media which will withstand the action of water (nonemulsifying 
property), air (nonoxidizing property), heat (noncarbonizing 
property), and cold (good cold test), and which, mixed with 
a small percentage of "very oily" constituents, are given the 
desired oiliness. 

In measuring devices, efforts are being made to construct a 
satisfactory commercial viscometer for measuring absolute 

The nature of friction and character of frictional surfaces are 
constantly being studied, and although this question still pre- 
sents many problems, considerable progress is recorded. 

In the mechanical field, an important development is the 
"Nomy" bearing principle, which appears to open up great 
possibilities for longer life and exceedingly low frictional losses for 
all types of bearings. 

A marked development has taken place inside the field of 
internal-combustion engines in the direction of better, cleaner, 



and more economic lubrication (e.g., modernized forms of piston 
rings, and scraper rings). 

The ever-widening use of circulation oiling systems for all 
high-speed engines and machinery has brought about an increased 
use of centrifugal oil purifiers and, in certain fields, streamline 
oil filters. 

August, 1937. 


Lubrication has for many years received only scant attention, 
and existing standards of lubrication still leave considerable 
room for improvement. Very few firms employ qualified chem- 
ists to assist them in maintaining a reasonable standard of 
efficiency; and such a thing as technical service embodying a 
highly trained staff of lubricating engineers was unheard of 
until recent years and is still considered an expensive luxury by 
most firms. 

A development is, however, gradually taking place in the 
right direction. Both oil suppliers and oil users are beginning 
to realize that lubrication can no longer be left to guesswork; 
that to send salesmen out with a set of samples and a price list 
but without the necessary technical knowledge or backing is 
to court failure; that entertaining customers or obtaining busi- 
ness simply through friendship between salesman and buyer 
is not sufficient, because friendship does not add to the lubricat- 
ing value of the oil, nor does it always help to select the right 
oil or use it in the right way. 

Lubrication is rapidly becoming a science. Some oil firms 
have appreciated the value of the assistance of a staff of qualified 
lubricating engineers, who should be able to inspect a plant, to 
report intelligently on the lubrication conditions of all engines 
and machinery, to point out and estimate the value of possible 
improvements in regard to savings in power or lubricants, to 
investigate complaints, etc. These men should have a thorough 
knowledge of their firms' products, so that they can recommend 
the correct grades for any kind of machinery, even without know- 
ing anything about the lubricants actually in use. 

Obtaining samples for analysis and " matching" them at a 
lower price per gallon is unfortunately still the standard of 
procedure of most oil firms and should be discouraged by the 
consumer in favor of a more efficient lubrication service, which 
places the supply of lubricants on a sound engineering basis. 

Large consumers of lubricants will find it worth their while to 
ask oil suppliers to demonstrate the value of their lubricants; 
they will soon find that it is of far greater importance than 



is generally realized that the lubricating systems of the engines 
or machinery be as perfect as possible; that the correct grades of 
lubricant be selected; that the lubricants be stored and dis- 
tributed in the best manner, used in the right way and in the 
right amount; and that the waste oil, if any, be collected, purified, 
and used again. 

Oil firms who intend to develop a technical organization must 
not make the mistake of thinking that they can engage any 
kind of engineers. A high standard of general engineering 
knowledge is essential, besides considerable tact in dealing with 

Furthermore, an engineer, however excellent his general 
knowledge, does not become a lubricating engineer the moment 
he is engaged by an oil firm. He will have to study the available 
literature but must not expect to develop his experience by 
sitting in the office. He should study closely lubrication of 
machinery under actual working conditions to the minutest 
details, and thus he will in time accumulate the right kind of 
special knowledge and develop the right instinct to enable him 
to render first-class service and to add his effort, be it great or 
small, to the advancement of the science of lubrication. 

The lubricating engineer needs good assistance from the 
chemical laboratory in analyzing oils, deposits, etc. On the 
other hand, chemists should not be expected nor should they be 
allowed to make recommendations, except in consultation with 
an engineer, who is able to investigate and judge the importance 
of the mechanical and operating conditions of the plant, which 
is essential in order to interpret correctly the value of the labo- 
ratory's findings. 

The oil manufacturer, through lubricating engineers, must 
watch constantly the results obtained under working conditions 
by the various standard grades of lubricants, and he will in this 
way accumulate knowledge as to the value and range of service 
of each particular grade; he will also find out possible weaknesses, 
and the engineering staff in conjunction with his chemical staff 
will be able to point the way to remedy. 

Oil firms who have developed an efficient technical staff will 
always have a great advantage over other firms who are less well 
equipped. Their salesmen having the benefit of technical 
assistance will easily command greater sales than their com- 


petitors. Even if their products are no better, they will yet be 
able to render to their customers better service, because they 
know how to select the correct grades and can indicate to the 
consumer how the maximum value of these grades can be 
obtained. Such service always brings credit and good will to the 
oil supplier and demonstrates to the consumer that lubrication 
service comprises a great deal more than is indicated by the 
price per gallon. 

The chief engineer or master mechanic of a works cannot be 
expected to know everything there is to know about lubrication; 
it is no discredit to him if he gains a few points by discussing the 
lubrication of his plant with lubricating engineers who have 
made a life study of the subject. 

The author hopes that oil firms who have no engineering staff 
will see the necessity of developing a technical service, sufficient 
for their needs in keeping with modern sales methods, which are 
directed toward selling lubrication, rather than lubricants, or 
selling experience and knowledge rather than selling oils on a 
price-per-gallon basis. 

The subject of lubrication is intimately connected with the 
mechanical and operating conditions of engines or machinery. 
The author has therefore endeavored to present for each type of 
engine or class of machinery the " technical background/' with- 
out which it is futile to attempt to focus the lubricating problems, 
as seen by the engineer or the chemist, and without which 
it is impossible to determine the character of the oils required 
to give the best service. 

The author is well aware of the magnitude of such a task and 
the many shortcomings of the present work, but he ventures to 
hope that the way in which he has dealt with the problems and 
endeavored to convey his experience may prove of some value 
in stimulating others to take a deeper interest in lubrication 
matters and in helping them to get a clearer view of possible 
problems or difficulties and their solution. 

Mechanical and electrical engineers in charge of plant and 
lubricating engineers as well as general consulting engineers will, 
the author hopes, find some food for thought; they may not 
always agree with the theories and views put forward, which are 
often novel or even contrary to traditional opinions; but in that 
case the author would urge them to try out his recommendations, 


which are based on many years of practical experience in many 
parts of the world; they will then be able to draw their own 
conclusions, and constructive criticism will always be welcomed 
by the author and gratefully received. 

Engine builders, it is hoped, will find information which will 
prove useful to them in equipping their engines and machinery 
with correctly designed lubricating systems and appliances 
and in giving their customers sound advice or instructions with 
reference to the grades of lubricants required and the best manner 
of using them. 

Oil chemists and manufacturers and chemists employed by oil 
consumers will, it is hoped, find the book helpful in pointing out 
the conditions under which lubricants have to work for particular 
types of machinery and the influences, such as oxidation, and 
emulsification, to which they are subjected during use. The 
author has endeavored to focus the problems and describe the 
mechanical conditions in such a manner as to assist chemists in 
deciding which are the physical and chemical tests of greatest 
importance in each particular case. 

References are given throughout the text to special sources of 
information, but the author wishes particularly to record his 
indebtedness to Mr. L. Archbutt for analyses of graphites; to Mr. 
J. Hamilton Gibson for photographs of streamlines in connection 
with Michell's thrust blocks; to Mr. I. L. Langton for information 
regarding dielectric strength of transformer oils; to The Engi- 
neer for permission to make use of some articles by the author 
on lubrication of modern turbines; to Mr. E. W. Johnston 
for information regarding the use of Aquadag in steam engines; 
to the Vacuum Oil Company of New York for raising no objection 
to the author's making use of several technical papers, which he 
prepared during the time he was associated with that company 
as chief engineer in London; to the Controller of His Majesty's 
Stationery Office for permission to make use of Bulletin 2 on 
cutting lubricants and cooling liquids and Bulletin 4 on solid 
lubricants, both of which have been published by the Department 
of Scientific and Industrial Research and the material for which 
was prepared by the author; and to Mr. W. A. E. Woodman for 
valuable assistance in preparing many of the drawings. 


August, 1920, 















































INDEX 629 





Oil Wells. Petroleum crudes are secreted by nature and are 
found in many countries all over the world. 

Occasionally, petroleum crude is found lying on the surface of 
water in pools, but usually it is found at various depths in the 
earth, from a few hundred feet up to as much as five thousand 
feet. To bring the crude to the surface, a hole is drilled, varying 
in diameter from a few inches up to 30 in., according to the 

It is usual to find confined with the oil a large amount of gas 
under great pressure, which may be as high as 800 Ib. per square 
inch. Owing to this pressure the oil when first reached is forced 
up the bore hole and rises many feet in the air; such a well is 
called a "gusher." 

Some gushers have produced enormous quantities of crude 
oil, for example, the Potrero No. 4 well drilled in 1910 by the 
Mexican Eagle Oil Company. This well was capable of giving 
about 120,000 bbl. of crude oil daily but now delivers only 
salt water. 

When the gas pressure is sufficiently reduced in an old well, 
it is no longer a " flowing well" but becomes a "pumping well," 
and the output is reduced to a small fraction of its former value. 

Production of Petroleum Crude. The production of crude oil 
amounts to approximately 300 million tons per annum. Of this 
production, 63 per cent is supplied by the United States, 16 per 
cent by Central and South America, 12 per cent by Europe, and 
9 per cent by Asia. 



United States. The production is still increasing in the United 
States; many of the old American fields (Pennsylvania, etc.) are 
becoming exhausted, but those in California and Oklahoma have 
made up for the decreased production in the older fields. 

Russia. Production has considerably increased after the 
World War, but exports have of late years decreased owing to 
largely increased home consumption. 

Mexico. The Mexican oil industry has developed rapidly 
since 1908. The potential resources are enormous, being prob- 
ably as great as or even greater than the resources of the United 

South America. New oil fields of considerable importance have 
been opened up in Peru, Venezuela, and Colombia. 

Persia. Large oil fields have been developed, and the output 
is increasing rapidly, particularly after building a large pipe line 
which connects the oil fields with the seacoast. 

Origin of Petroleum Crude. Three theories are held concern- 
ing the origin of crude oils, but none is universally accepted. 

1. Inorganic Theory, According to this theory, petroleum is 
produced deep down in the crust of the earth by the action of 
high temperature and pressure on the minerals found there; 
carbon and hydrogen are supposed to have combined and formed 
the hydrocarbons which are the chief constituents of petroleum 
crude. Only a minority of geologists favor this theory. 

2. Vegetation Theory. According to this theory, vegetable 
matter has been covered by a layer of impervious material; the 
air thus being excluded, rotting was prevented, and slow decay 
during hundreds of thousands of years transformed the vegetable 
matter into petroleum crude oil and petroleum gas. Several 
geologists favor this theory. 

3. Marine-animal Theory. According to this theory, dead 
fishes or tiny marine animals with calcareous shells were covered 
over by a layer of impervious material, and their organic parts 
have gradually been transformed into crude oil and gas. Most 
geologists favor this theory. 

Whichever theory is correct, it seems certain that the world's 
stocks of petroleum crude are practically complete and are being 
rapidly consumed. 

Composition and Character of Petroleum Crude* When the 
crude comes to the surface it often contains water (frequently 


salt water) and sand, which are separated out in large collect- 
ing and settling reservoirs. 

The crude is rarely transparent; the color is usually dark 
brown or black. 

Petroleum crude consists chiefly of carbon (C) and hydrogen 
(H) in the form of hydrocarbons. Besides carbon and hydrogen, 
there is usually also a certain amount of oxygen, nitrogen, and 
sulphur present. 

The percentages of the various chemical constituents vary 
within limits, as indicated in the following table: 

Per Cent 

Carbon 81.00 to 88.0 

Hydrogen 10.00 to 14.0 

Oxygen 0.01 to 1.2 

Nitrogen 0.002 to 1.7 

Sulphur 0.01 to 5.0 

Hydrocarbons. Paraffins (C n H 2n + 2 ). The molecules of these 
hydrocarbons are bound together in the form of chains and are 
members of the large family of hydrocarbons known as open- 
chain hydrocarbons, thus: 

H H H H H H 

II II I i 

H C C C C C C H 

H H H H H H 

As all the carbon atoms are fully engaged, each carbon atom 
being tetravalent and attached to four other atoms, the paraffins 
are called saturated hydrocarbons. 

Olefins (C n H 2n ). The olefins are also open-chain hydrocarbons, 
but their molecules have two atoms of hydrogen less than the 
paraffins, thus: 


H C C C C = C C C C H 


They are called unsaturated, because they are capable of absorb- 
ing hydrogen, oxygen, sulphur, etc., to a value equivalent to 2 
atoms of hydrogen per molecule. 


Napkthenes (C n H 2n ). Naphthenes are closed-chain hydrocar- 
bons; they have the same chemical formula as the olefins, except 
that the atoms are not arranged in the form of open chains but 
more in the nature of rings or closed chains, in such a manner as 
to saturate all the carbon atoms. 

Naphthenes being saturated hydrocarbons are consequently 
more stable than the olefins. 

Cn#2n-2, C n #2n-4, Etc. Most hydrocarbons of the formulas 
C n H 2n _ 2 , C n H 2n _ 4 , etc., are more or less unsaturated, and the 
more so the less hydrogen they contain. 

Hydrocarbons having from 1 to about 15 carbon atoms per 
molecule represent the light products of petroleum crude, viz., 
petroleum gas, gasolines, kerosenes, and light transformer and 
spindle oils. 

Most lubricating oils are mixtures of hydrocarbons possessing 
more than 15 carbon atoms per molecule; the greater the number 
of carbon atoms the greater is the viscosity of the oil. Comparing 
two hydrocarbons having the same number of carbon atoms, the 
one containing the least hydrogen is the more viscous of the two, 
but its viscosity is less stable; i.e., it changes more rapidly with 
changes in temperature. 

Most petroleum crudes are very complicated in character, and 
it is difficult to classify them; they contain hydrocarbons of 
practically all types, but the proportions vary considerably 
according to the origin of the petroleum. 

Petroleum crudes are, however, referred to as paraffin-base 
crudes, naphthenic crudes, asphaltic-base crudes, and mixed-base 

Paraffin-base crudes are so called because they contain 
paraffin hydrocarbons (C n H 2n4 . 2 ). There are only a few lubri- 
cating oils of low viscosity which are actually paraffin hydro- 
carbons, as paraffins from CirHae and upward represent the 
hydrocarbons present in paraffin waxes. The heavy-viscosity 
lubricating oils which are found in paraffin-base crudes are 
largely composed of olefins and naphthenes (C n H 2n ) and acety- 
lenes (C n H 2n _ 2 ). As paraffin-base crudes always contain a 
certain amount of paraffin wax, usually about 2 per cent, lubricat- 
ing oils made from such crudes have high setting points. 

The most important supplies of paraffin-base crudes come from 
Pennsylvania and Ohio in the United States. They are fairly 


fluid; rich in gasolines and kerosenes; usually contain only a 
little asphalt, sulphur, oxygen, or nitrogen; and have a low spe- 
cific gravity. 

Naphthenic crudes consist chiefly of naphthenes (C n H 2n ) ; they 
also contain a small percentage of acetylene hydrocarbons 
(C n H 2n _ 2 ). 

Russian and some South American crudes belong to this class 
and contain little or no paraffin wax, hence produce lubricating 
oils with low setting points. 

Asphaltic-base crudes are so called because they contain a 
large amount of asphalt; they usually contain certain small per- 
centages of sulphur, oxygen, and nitrogen. The crudes from 
California, Mexico, Texas, and South America belong to this 

They are very viscous; black in color; rich in lubricating oils, 
fuel oils, and asphalt; and have a high specific gravity; they often 
contain complex sulphur compounds, which are difficult to extract. 

The lubricating oils produced from nonparaffinic asphaltic-base 
crudes are naphthenic in character, have low setting points, and 
possess a wide range of viscosity, ranging from quite thin to 
exceedingly viscous oils. 

The hydrocarbons in asphaltic-base crudes are lower in hydro- 
gen than the paraffins, although paraffins are often present. 
For example, California crudes contain olefins (C n H 2n ), asphaltic 
hydrocarbons (C n H 2n _ 4 ), and also some aromatic compounds 
(C n H 2n __ 6 ). 

Mixed -base crudes are crudes of a character intermediary 
between paraffin-base crudes and asphaltic-base crudes, contain- 
ing both paraffin wax and asphalt. Mexican crudes are the most 
important mixed-base crudes. 


Petroleum crude is a mixture of many hydrocarbons, all having 
different boiling points. 

The separation of all these hydrocarbons frpm the lightest 
gasoline to the heaviest lubricating oils is done by means of 
distillation and subsequent condensation of the various "frac- 
tions." The old intermittent system by which batches of, say, 1,000 
bbl. of crude oil were distilled in individual stills has rapidly been 
replaced by the modern continuous system of distillation. In this 


system, the crude oil is pumped through nests of tubes which are 
heated by oil fire. Notwithstanding that the crude oil passes 
very rapidly through the tubes, it becomes heated to such a high 
temperature that the latent heat is about sufficient to evaporate 
it completely. When it next passes over " flashes' 7 in one 
or several towers, broken dephlegmation of the vapors takes 
place, while the nonevaporated residue collects and may be drawn 
out at the bottom of the tower. 

In order to regulate the dephlegmation, it is customary to 
introduce superheated steam or kerosene vapors at one or more 
places in the tower. 

Cracking. The crude itself or certain distillates are " cracked " 
when it is desired to produce the maximum amount of light frac- 
tions. When hydrocarbons are suddenly heated to a tempera- 
ture above their boiling points and not given time to distill in the 
ordinary way, they decompose iiito simpler hydrocarbons which 
possess lower boiling points; this process is called " cracking. " 

Cracking is now practiced according to two different principles : 

1. Liquid Cracking. The heavy oil is rapidly heated in the 
liquid state the lighter constituents evaporate the residue is 
again heated, etc., until the maximum amount of light oil is 

2. Vapor Cracking. The heavy oil is evaporated; the vapors 
are led through a reaction chamber in which they are exposed to 
high temperatures. The light fractions thus produced are 
separated out by dephlegmation, while the heavier fractions are 
again evaporated and sent through the reaction chamber. 

Most cracking methods now employed operate according to 
the foregoing principles or a combination of these. Often the 
process finishes by refining, the oil vapors passing through fuller's 
earth or other refining chemicals. 

By cracking, a large amount of aromatic and unsaturated 
hydrocarbons are formed; aromatic compounds in gasoline mean 
high " antiknock value " (a high octane index), which makes such 
gasolines specially suitable for use in high-compression gasoline 

Steam Distillation. When it is desired to produce the maxi- 
mum amount of lubricating oils and minimize cracking, live super- 
heated steam is introduced into the stills, mixing intimately with 
the oil. 


To increase further the yield of lubricating oils, and to prevent 
overheating, the oil may be distilled under a partial vacuum, as 
the vacuum causes the various fractions to distill over at tem- 
peratures lower than their normal boiling points. 

When the distillation is assisted by the application of steam 
with or without vacuum, a lower percentage of unsaturated 
hydrocarbons is formed than when distilling without steam, and 
less acid or other treatment is therefore required when refining 
the distillates. 

Redistillation. Usually, the crude is split into only a few 
fractions, which may be further separated into a greater number 
of fractions by redistillation. 

Lubricating distillates are also redistilled and thus separated 
into heavier and lighter lubricating oils. The redistillation now 
nearly always takes place in towers under high vacuum and with 
application of superheated steam. With certain systems a 
vacuum as high as 0.1 mm. Hg is employed. 

Extraction. During late years, methods have been devised 
whereby lubricating oils are extracted direct from the crude oil, 
e.g., by mixing with propane and phenol. The liquid propane 
separates out the asphaltic matter, and the phenol extracts the 
aromatic and naphthenic constituents, leaving an oil more or less 
paraffinic in character, according to how much of the aromatic 
and naphthenic contents is extracted. 

By these methods, even a Texas crude may be divided into 
fractions of naphthenic as well as paraffinic character. 


When the light fractions, viz., gasolines (distilling up to 150C.) 
and kerosenes (distilling between 150 and 300C.), have been 
distilled off, the next distillate is a high-flash burning oil called 
"300 fire-test " oil, mineral colza, mineral sperm, or mineral seal; 
but if the quality of this distillate is not such as to produce a 
satisfactory burning oil, the distillate is called " solar oil" or "gas 
oil" and is used for making oil gas or carbureted water gas or as a 
high-class fuel oil for semi-Diesel or Diesel oil engines. Also, 
when mixed with heavy black residual oils (asphaltic or non- 
asphaltic) it is used as fuel oil in Diesel engines or in furnaces 
using liquid fuel. 


The lubricating-oil fraction or fractions (from which spindle 
oils, engine, and machinery oils are manufactured) now distill 
over; and if the crude contains wax, the distillate containing wax 
is chilled to about 20 to 25 F., and in the wax filter press the oils 
are squeezed out, and the wax left in the press. Lubricating oils 
made from a paraffin-base crude, therefore, have setting points 
of about 20 to 25F. unless they are specially treated to remove 
more of the wax; they may also be blended with other oils having 
very low setting points so as to produce oils with low setting 

In modern refineries, the wax is often removed by dissolving 
the oil in gasoline or chlorinated hydrocarbons, cooling the solu- 
tion, and subsequently separating the wax in special centrifugal 
separators. When employing gasoline, it is possible to remove 
only the amorphous wax continuously, but with chlorinated 
hydrocarbons, which are heavier than the oil, it is possible con- 
tinuously to separate crystalline as well as amorphous wax. 

After centrifuging, the solvent is distilled away, and an oil 
remains with a cold test which bears a relation to the temperature 
to which the mixture was cooled and to the amount of solvent 

The cold test of an oil may also be reduced by adding to it 
so-called inhibitors, e.g., Paraflow. The effect of such inhibitors 
is that, when the oil is cooled, the wax solidifies or crystallizes in 
the form of very small particles which do not touch each other 
but merely float about, so that the oil remains fluid. 

The crude wax, whether produced by filter presses or by 
centrifuging, contains a certain amount of oil (up to 50 per cent), 
which is removed by "sweating, 11 viz., slow prolonged heating of 
the wax. The melting points of the sweated wax range from 100 
to 130F.; it is melted, crystallized in molds, and sold as white 
paraffin wax used chiefly for making candles, also for preserving 
fruit and jellies, for polishing floors, etc. 

The pressed or centrifuged lubricating oils are redistilled into 
heavier and lighter oils and either (1) treated by sulphuric acid 
or anhydrous aluminium chloride or (2) filtered through fuller's 
earth, or high-activated fuller's earth, in order to remove unstable 
hydrocarbons or other undesirable elements and to lighten the 


When filtered through fuller's earth or animal charcoal, the 
first few gallons of oil that come out are colorless; but as the 
filtering material becomes saturated with the absorbed impurities 
and coloring matter, the color of the oil gradually darkens. 
Each grade of oil is filtered to be within the standard color limits 
for that particular grade. 

Dark Cylinder Stock. The residue of some crude oils from 
distillation is a very heavy viscous dark oil used principally for 
internal lubrication of steam-engine cylinders and valves. If it 
contains much more than a trace of asphalt, it should not be used 
for this purpose but may be mixed with light-viscosity lubricating 
oils to produce dark lubricating oils. 

Filtered cylinder stock is produced from dark cylinder stock 
by filtration; the color becomes green-amber; the heavy-gravity 
tarry matter is removed; the viscosity is reduced 15 to 25 per 
cent; and the specific gravity is likewise reduced, but the setting 
point is increased. 

Bright stock indicates normally a filtered cylinder stock, from 
which wax has been specially removed, but there are also a good 
many very viscous distillates on the market which are sold under 
this name. 

Petroleum jelly (mineral jelly, petrolatum) is an amorphous 
wax produced by slow cooling of dark cylinder stock diluted with 
gasoline; the petroleum jelly separates out and is afterward refined 
(decolorized) by hot filtration. Petroleum jelly is used in the 
manufacture of cordite (an addition of 2 per cent of jelly makes 
the cordite less brittle), as an antirust grease, for ointments 
(veterinary purposes), etc. Vaseline is the proprietary name 
given to a certain high-grade petroleum jelly. 

Cold-test Cylinder Stock. By distilling off the gasoline from 
the liquid portion a low cold-test cylinder stock is produced, which 
may be further refined by filtration. 

The best cylinder stocks are almost exclusively produced from 
paraffin-base crudes. 

When asphaltic-base crudes are distilled, cylinder stock can 
rarely be produced; the residue consists of a&phaltic matter. 
Heavy liquid asphaltic residues are used as road-spraying material 
in place of coal tar and are also used in the manufacture of 
various liquid fuels. 


Petroleum pitch or bitumen has found a most important use, 
chiefly in the making of wearing surfaces for modern roads, also 
'or roofing felts, bituminous paints, etc. It is also used in the 
naking of hot-neck greases for steelworks rolling mills. 

When the liquid bitumen in the stills is "blown" with air, it 
oxidizes into blown asphalt, which has a rubbery nature and finds 
in important use as rubber substitute, for roofing felt, etc. 


Dark Cylinder Oils. Dark cylinder oils are the undistilled 
lark residues left in the stills (by steam distillation chiefly of 
lonasphaltic crude), freed from solid impurities but not filtered. 
They are used chiefly for lubrication of steam-engine cylinders 
md valves, either alone or mixed with from 3 to 10 per cent of 
icidless tallow oil. The ordinary characteristics are as follows: 

Flash point open From 500 to 620F. 

Specific gravity From 0.900 to 0.940 

Viscosities Nos. 11 to 16 (see page 57) 

Dolor in reflected light Dark brown or dark green to black 

Dolor in transmittent light Dark brown to black 

Setting point 25 to 60F. 

Filtered Cylinder Oils and Bright Stocks. They represent the 
lighest quality oils used for internal lubrication of steam engines; 
:hey are used either alone or mixed with from 3 to 12 per cent of 
icidless tallow oil. They are also used largely for mixing with 
ower viscosity oils to produce heavy-viscosity oils for internal- 
combustion engines or heavy-viscosity engine and machinery oils, 
air-compressor oils, circulation oils, etc. 

Flash point open From 490 to 580F. 

Specific gravity From 0.875 to 0.930 

Viscosities Nos. 11 to 15 (see page 57) 

Color in reflected light Green, amber 

Color in transmittent light Deep red 

Setting point 15 to 80F. 

Ordinary lubricating oils are distilled or extracted and then 
refined and filtered. 

The heavier engine or machinery oils may also be produced by 
mixing the lighter oils with filtered cylinder oil or bright stock. 


Ordinary lubricating oils represent the great bulk of the oils 
used for general external lubrication of all kinds of engines and 

Every oil refinery of importance may be relied upon to produce 
lubricating oils for such general purposes as fulfill all reasonable 
ordinary chemical and physical requirements, such as viscosity, 
cold test, flash point, freedom from acidity, etc. 

One may divide this group into two typical groups, viz., spindle 
and light machinery oils and heavy engine and machinery oils, 
having the following characteristics: 

Spindle and Light Machinery Oils. They are light to medium 
in viscosity and are used for quick-running machinery, such as 
textile machinery, high-speed shafting, electric motors; and also 
for manufacturing yellow lubricating greases: 

Flash point open 275 to 420F. 

Specific gravity 0.870 to 0.910 

Viscosities Nos. 1 to 8 (see page 57) 

Color Pink to red 

Setting point (paraffin base) 15 to 25F. 

Setting point (asphaltic base) to 15F. 

Heavy Engine and Machinery Oils. They are of high viscosity 
and used for slower running engines and machinery and for heavier 
bearing pressures. 

When mixed with from 5 to 20 per cent of fixed oil (blown or 
unblown) they produce some of the lighter viscosity marine- and 
railway-engine oils: 

Flash point open 380 to 440F. 

Specific gravity 0.900 to 0.930 

Viscosities Nos. 9 to 12 (see page 57) 

Color Red 

Setting point (paraffin base) 20 to 30F. 

Setting point (asphaitic base) to 20F. 

High-grade lubricating oils are needed for such purposes as 
circulation lubrication of steam turbines and highspeed enclosed- 
type steam engines, internal lubrication of air compressors, 
refrigerator compressors, all kinds of internal-combustion engines, 

Great knowledge and experience are required on the part of the 
oil refiner to produce oils for such exacting requirements and to 


keep pace with the never ending development of modern engines 
and machinery. 

The oils are exposed to such influences as extreme heat or cold, 
oxidation, emulsification, electric action, all of which will be 
discussed later when the various types of modern engines and 
machinery are described. 

The difference between ordinary and high-grade oils lies chiefly 
in that the latter are the outcome of extreme care all the way from 
selecting the crude to the final treatment, with a view to giving 
the oil just those special properties which are desired for the 
particular service in question. 

It is obvious that no oil refinery can manufacture high-grade 
oils without a full knowledge of the conditions under which high- 
grade oils are expected to operate and that, in consequence, there 
must be close cooperation between the refinery chemists and the 
service engineers. 

Dark Lubricating Oils. Dark lubricating oils are such undis- 
tilled residues from the crude or from the redistillation of lubricat- 
ing oil distillates that, because of too low a viscosity or for other 
reasons, are considered unsuitable as cylinder oils. Dark 
lubricating oils are usually mixtures of such residues with low- 
viscosity lubricating oils to produce the required viscosity. 

Dark lubricating oils are used for rough machinery in collieries 
and steelworks, as cheap oils for lubricating the axles of railway 
carriages, and for making black lubricating greases for rough 

Flash point open 300 to 450F. 

Specific gravity 0.890 to 0.960 

Viscosities Nos. 10 to 13 (see page 57) 

Color Dark green or brown to black 

Setting point 10 to 60F. 

Asphalt Less than 5 per cent 

Bloomless Oils. Bloomless oils are neutral oils that have been 
highly filtered (not acid treated) and may also have been sun 
bleached; they are very light in color and of light viscosity. 

To have the bloom entirely removed, they must be treated with 
nitronaphthalene or other chemicals. 

Bloomless oils are used for adulterating edible oils; also in the 
manufacture of " stainless " loom and spindle oils. 


White Oils. White oils are pale spindle oils which have been 
treated with fuming sulphuric acid or liquid sulphur dioxide, 
fuller's-earth filtration, etc., in order to remove the color com- 
pletely. They are easily made from Russian crudes and are 
largely used as transformer oils. It is very difficult to remove 
color entirely from oils produced from paraffin-base crudes. 

Medicinal White Oils. Medicinal white oils are white oils 
that have been so treated as to remove not only color but also 
all taste and odor. 



Vegetable Oils and Fats Animal Oils and Fats 

Castor oil Tallow 

Rape oil Tallow oil 

Blown rape oil Lard oil 

Cottonseed oil Neat's-foot oil 

Blown cottonseed oil Sperm oil 

Linseed oil Whale oil 

Olive oil Porpoise oil 

Coconut oil Dolphin oil . 

Palm oil Melon oil 

Palm-kernel oil Menhaden oil 

Peanut oil Cod oil and other fish oils 

Mustard oil Wool grease 
Rosin oil 

Animal and vegetable oils are called " fixed" oils because they 
cannot, like mineral oils, be distilled without decomposition. 
They also differ from mineral oils in that they contain from 9.4 to 
12.5 per cent oxygen. 

The distinction between fixed oils and fats is only a matter of 
temperature; all fixed oils become fats at or above 0F., and all 
fats become oils at or below 125F. 

Animal oils are obtained by heating the fatty tissues of animals, 
i.e., by " rendering" the fat or by boiling out the fatty oil with 
water. Vegetable oils occur mostly in the seeds or fruits of 
plants or trees and are obtained either by pressing or by chemical 
extraction with solvents. Animal oils are usually either colorless 
or yellow. Vegetable oils are colorless, yellow, or slightly green 
(chlorophyll present). 

All fixed oils are devoid of bloom except rosin oil, and each 
variety generally has a distinctive odor, by which it can be identi- 
fied. Their specific gravities range from 0.860 to 0.970. Rosin 
oil is an exception; its specific gravity may be as high as 1.0. 
Sperm oil has the lowest viscosity of all fixed oils, and castor oil 
the highest, but each kind of oil has its own peculiar viscosity, 
which varies only slightly. 




All fixed oils have a tendency to combine with oxygen and, as 
a result, are sooner or later converted into solid elastic varnishes. 
As a result of this tendency, cotton waste, when saturated with 
fixed oils or lubricating oils very rich in fixed oils, has been known 
occasionally to heat gradually and finally to burst into flame. 
Dirty cotton waste, which contains fixed oil, must therefore be 
kept in receptacles with closed lids. 

When the tendency to absorb oxygen is marked, the fixed 
oils are called drying oils, e.g., linseed oil. When the tendency 
is moderate or only slight, the oils are called semidrying or 
nondrying oils, respectively, and it is only from these two types 
of fixed oils that lubricants are selected. 

Mineral lubricating oils are practically free from any tendency 
to oxidize and therefore do not readily gum or develop acid as 

(Journal of Society of Chemical Industry, Vol. XVIII, p. 346, 1899.) 


Name of acid 


Occurs chiefly in 


C 6 H 10 O 2 

Porpoise oil 


C H O 

Caprylic . . . 

C R 1 O 



C H O 

Coconut oil 

C Ho O, 


C H O 




Palm oil, also tallow, olive 


C H 

oil, and coconut oil 
Tallow, also palm, castor, 

Arachidic . . 

v_y 20-ti 40^- ' 2 


and rape oils 
Earthnut, rape and mustard 





Cl8H 34 O2 

Olive oil and the animal 


C Hon 9 O 2 

Rapic .... 

V_y i sJtl 34^-' 2 





Rape oil 




C H 2 O 

The drying oils, also in olive 

C n H 2n _ 4 O 2 

and palm oils 



\~/ 1 8-ti-34v-' 3 


C H^ 9 O, 


Castor oil 

Q,H*o 4 

Dihydraxystearic . . 

C 18 H M 4 

Castor oil 


fixed oils do, which may lead to corrosion of the bearing 

All fixed oils are chemical combinations of alcohol radicles and 
fatty-acid radicles. The character of fatty acids is indicated 
in the table on page 15. The alcohol radicle occurring in the vege- 
table oils and most of the animal oils is glyceryl (CaHs), which is 
trivalent and therefore combines with three fatty-acid radicles. 
Olein, for example, which is the chief constituent of many fixed 
oils, such as tallow, lard, neat's-f oot and olive oils, has the chemi- 
cal formula CaHsCCisHaaO^a, in which C, H, and O signify 
carbon, hydrogen, and oxygen atoms, respectively. Stearin 
[Cal^CiaHasC^al and palmitin [CaHsCCieHaiC^a] predominate in 
solid fats; olein, in the fluid oils. It will therefore be seen that 
the nature of the fatty-acid radicle determines the character of 
the fixed oil. 

Sperm oil is made up differently, being known as a liquid wax. 
All fixed oils, however, can be split up into alcohols and fatty 
acids by heating with water under pressure, by heating with 
sulphuric acid, by heating with alkalies, etc. By treatment with 
alkalies the fixed oils are said to be saponified. For example, by 
heating olein with water under pressure, the following change 
takes place: 

= CaHUCO.H^ + 

Olein Water Glycerin Oleic acid 

This change takes place in steam cylinders, when too high a 
percentage of fixed oil is used in the cylinder oil; the fatty acids 
thus liberated eat away the metal and form metallic soaps. 

By heating olein with an alkali, e.g., potash, the following 
change takes place: 

(C 18 H33O 2 )3 + 3KOH = C 3 H B (OH) 3 

Olein Potash Glycerin Potash oleate 

It will be seen that the fatty acid is not now liberated but has 
combined with the potash and formed a soap. 

This action distinguishes fixed oils from mineral oils, which 
are not saponified when heated with an alkali but remain 

(See also Tables, pages 23 and 24) 

Vegetable Oils and Fats. Castor Oil (Nondrying). Castor 
oil is obtained from the seeds of the castor tree or shrub, which 


grows in all tropical and subtropical countries. The kernel 
forms 80 per cent of the seed and yields about 50 per cent of 
its weight in oil. By cold pressing of the seeds, medicinal castor 
is produced. By hot pressing " first pressings" and "second 
pressings" are afterward produced. Castor oil may also be 
extracted by solvents. Crude castor oil is refined by steaming 
and filtration. When properly refined, castor oil keeps well and 
does not easily turn rancid. 

Castor oil is liable to deposit a solid fat in very cold weather 
but congeals only at very low temperatures. It is nearly colorless 
or slightly greenish yellow; it has the highest specific gravity 
and viscosity of all fixed oils; it is soluble in alcohol but not in 
petroleum spirit when cold, nor does it mix to any large extent 
with mineral oils. It mixes with refined rosin oil in all propor- 
tions. It will absorb a maximum of about 12 per cent of pale, 
low-setting-point mineral lubricating oil, whereas mineral oil 
will not absorb much more than 3 per cent of castor oil. 

All fixed oils, except castor, mix readily with mineral oils, and 
it is quite easy to make clear mixtures of castor oil and mineral 
oil in the presence of another fixed oil, such as lard oil or rape oil. 

Castor oil is an excellent lubricant, possessing great oiliness. 
It is used for lubricating bearings subjected to great pressure, 
such as heavy-type marine engines, and is extensively used for 
airplane engines, particularly the rotary types, which cannot be 
lubricated satisfactorily with any oil other than pure medicinal 
castor. It is also used in the manufacture of soluble oils, in the 
manufacture of greases for pistons with India-rubber or leather 
fittings, as a preservative for rubber and leather belting, etc. 
The possibilities of castor oil as a lubricant appear to be far 
from exhausted. For example, little work has been done with 
blown castor oil, nor does there appear to be any satisfactory 
method developed for making miscible castor oil. One method is 
to heat castor oil for a few hours at 4 to 5 atmospheres pressure ; 
this treatment changes its nature and makes it more miscible 
with mineral oil. 

Treated with sulphuric acid, castor oil takes up 25 per cent 
of water and becomes "Turkey-red" oil used in preparing cotton 
fiber for dyeing. 

Rape Oil (Colza) (Semidrying) . Rape oil is obtained either 
by expression or by extraction from rapeseed, grown chiefly in 
India and Russia. Crude rape is dark in color and contains 


slimy impurities which are removed by treatment with sulphuric 
acid, followed by agitation with steam and hot water. If not 
sufficiently treated with acid, the slimy impurities choke the 
lubricating grooves; it is preferable to prolong the acid treatment 
and make sure of the elimination of the impurities, notwithstand- 
ing the development of a little extra free fatty acid. 

Black Sea rape oil (Ravison rape) is expressed from seeds of 
the wild rape of the Black Sea district; it is inferior to ordinary 
rape oil, being about 10 per cent lower in viscosity and having 
a greater tendency to oxidize (more " drying"). 

Blown rape oil is rape oil that has been blown with air at a 
temperature rising during the process from 160 to 250F. The 
oil is oxidized, increases greatly in specific gravity and viscosity, 
and develops free fatty acid. The specific gravity may be 
increased from 0.915 to as much as 0.985. 

When rape oil is blown, the color darkens for about 3 hr.; 
then the oil becomes pale, but at the finish of the operation it 
darkens to a deep red; it gives off considerable odor, but the 
finished oil has no odor. The viscosity at first decreases corre- 
spondingly with the pale color, then increases, becoming 200 sec. 
Saybolt at 212F. after 22 hr., 720 sec. Saybolt after 34 hr., etc. 

Rape oil or blown rape oil is chiefly used in the manufacture 
of railway- and marine-engine oils, from 10 to 25 per cent being 
mixed with heavy-viscosity (preferably low-setting-point) mineral 
oils at a temperature of about 140F. Rape oil is also used in the 
manufacture of soluble oils and as a quenching oil for steel. 

Rape oil mixes in all proportions with mineral oil, but with 
blown rape oil there is a minimum percentage below which the 
blown rape will not mix with the mineral oil. This minimum 
percentage is less at lower temperatures, so that sometimes in 
cold weather the blown oil separates out. The blown oil also 
separates out, if oil containing blown rape is diluted sufficiently 
with mineral oil. 

Cottonseed Oil (Semidrying) . Cottonseed oil is obtained by 
expression from cotton seed. On account of its drying properties, 
it should not be used for lubrication; it is, however, often used to 
adulterate olive oil, rape oil, or lard oil. Blown cottonseed oil is 
used as a substitute for blown rape oil in the manufacture of 
marine-engine oils but is not to be recommended. As a cutting 
oil it is used to give a high degree of " finish." 


Linseed Oil (Drying). Linseed oil is obtained from the seed 
of flax, is pale yellow in color, and is the best known of the drying 
oils. It cannot be used as a lubricant. 

Olive Oil (Nondrying). Olive oil is obtained by expression 
from the fruit of the olive tree. Fine olive oils are pressed cold 
and are used as salad oils as well as for lubrication. Olive oils 
from the second pressing (hot) are used for lubrication but are 
inferior to cold-pressed olive oil; they are more inclined to "dry," 
contain a rather high percentage of free fatty acid, and easily 
become rancid. Olive oils have now practically gone out of use 
for lubrication, having been displaced by mineral oils or mixtures 
of such oils with rape oil. 

Olive oil is largely used as wool oil in the high-class woollen 
industry; it is unsurpassed for this purpose, lubricating the 
woollen fibers during manufacture and being completely scoured 
out of the yarn when completed. It is used for lubricating high- 
quality cloth looms or finishing machines, as, if it gets on to the 
cloth, the stains disappear entirely in the scouring process. 

Coconut Oil (Nondrying). Coconut oil is produced from coco- 
nuts, the fruits of a certain kind of palm tree. The kernels 
are cut up and dried in the sun, producing the so-called " copra " 
from which coconut oil is obtained by expression. 

Coconut oil is fluid in tropical climates and solid in colder 
climates, the melting point being 70 to 80F. By cold pressing 
a fluid, coconut olein, is obtained which is used for lubricating 
purposes; the solid portion is used as an edible fat. 

Coconut olein is used to the extent of from 3 to 10 per cent in 
the manufacture of oils for internal-combustion engines. 

Palm Oil 9 Palm-kernel Oil (Nondrying). Palm oil and palm- 
kernel oil are obtained from the fruit of the African oil palm. 
The palm oil is produced from the fleshy layer, or pericarp, 
surrounding the hard woody shell, within which is the seed kernel. 
The palm-kernel oil is produced from the kernels and is quite 
different from palm oil; it closely resembles coconut oil but usually 
contains a large proportion of free fatty acid and is not used for 
lubrication. , . 

Palm oil varies in color from yellow to deep red; the odor is 
pleasant; the melting point ranges from 80 to 110F., the higher 
values corresponding with high percentages of free fatty acid, 
which are present to the extent of 10 to 40 per cent or even more. 


Palm oil is used in the manufacture of railway lubricating 

Peanut Oil, Also Called Earthnut Oil, Groundnut Oil, Arachis 
Oil (Nondrying). This oil is obtained from the nuts of a creeping 
plant called Arachis hypogcea. It is pale greenish yellow in color, 
of a nutty flavor and odor, but is now made nearly colorless and 
tasteless for edible purposes. It contains about 5 per cent of free 
fatty acid and is a nondrying oil. Peanut oil is used in the same 
manner as coconut olein is the manufacture of oils for internal- 
combustion engines. 

Mustard-seed Oil. Mustard-seed oil is said to have lubricating 
properties similar to those of castor oil, but it does not appear to 
have been much used as yet for lubrication. 

Rosin Oil (Semidrying) . Rosin oil is produced by destructive 
distillation of colophony (common rosin). The first products 
distilling over are rosin spirits. The rosin oil distills over above 
300C. (572F.) and may amount to 85 per cent of the total 
products. The residue in the still is rosin pitch or, if the distilla- 
tion is carried to dryness, coke. 

Crude rosin oil is a brown, viscous liquid with a strong blue or 
violet fluorescence. By heating to 150C. for three or four hours 
the fluorescence changes to green, and it loses from 1 to 5 per cent 
of its more volatile constituents. It contains a considerable 
percentage of rosin acids. 

Pale rosin oils can be produced by refining the crude rosin oil. 
The bloom can be removed by sun bleaching in shallow vessels 
or by treatment with nitronaphthalene, hydrogen peroxide, etc. 

The specific gravity ranges from 0.96 to 1.01. 

Rosin oil is not used as a lubricant in the ordinary way, but 
both rosin and rosin oil are successfully used in the manufacture 
of soluble oils, belt dressings, etc. It is also used in the manu- 
facture of low-quality lubricating greases. 

Animal Oils and Fats. Tallow (Nondrying). Beef tallow is 
obtained from cattle; mutton tallow, from sheep and goats. In 
rendering tallow for lubrication, it is important to use only fresh 
fat, which has not become decomposed, and to remove by settling 
and straining all water and membrane. 

Tallow from 60 to 80F. is a mixture of solid and fluid fate. 
When used for lubrication it should preferably not contain 
more than 4 per cent of free fatty acid in terms of oleic acid. 


Beef tallow is less inclined to become rancid than mutton 

Tallow is used in the manufacture of white tallow greases, also 
in most other lubricating greases to form the saponified base 
which "holds" the lubricating oil in the grease. Unrendered 
tallow suet is sometimes used for lubricating badly worn, 
open-type bearings. 

Tallow Oil (Nondrying). If tallow is subjected to pressure, 
the liquid portion can be separated out and is known as tallow 
oil. Acidless tallow oil is carefully made tallow oil and is used 
chiefly in the manufacture of steam cylinder oils, the admixture 
of tallow oil being from 3 to 15 per cent. It is also used in the 
manufacture of cutting oils. It should have a low content of 
fatty acid and a clean sweet odor; it should be colorless or pale 
yellow and free from suspended matter. 

Lard Oil (Nondrying). Lard oil is a fluid oil expressed from 
pig's fat. Winter-pressed lard oil has a lower setting point than 
summer-pressed lard oil. The setting point depends entirely 
upon the temperature at which the oil has been pressed; it may 
range from 32 to 60F. 

Prime lard oil is nearly colorless or pale yellow. 

Tinged lard oil is a second-quality lard oil, being more or less 
colored (yellow to brownish red) and containing a high percent- 
age of free fatty acid from 8 to 15 per cent or more. 

The best grades of lard oil are used in the manufacture of 
cutting oils (5 to 100 per cent lard oil), in the manufacture of 
internal-combustion engine oils (3 to 10 per cent lard oil), also in 
the manufacture of stainless oils. Tinged lard oil is nearly always 
used instead of prime lard oil in making cutting oils, but not in a 
greater proportion than 15 to 25 per cent on account of its bad 
odor and a gumming tendency greater than that of prime lard. 

Neat's-foot Oil (Nondrying). Neat's-foot oil is obtained by 
boiling the hooves and bones of cattle in water and skimming off 
the oil from the surface. When the oil is chilled and pressed, a 
low-setting-point neat's-foot oil is produced, which is much used 
for lubrication of watches and scientific instrument^; it is used 
for lubricating the air-operated engines in torpedoes, also for 
lubricating lacemaking machinery on account of its clinging and 
stainless properties. The high price of neat's-foot oil has con- 
fined its use as a lubricant to such special purposes. 


Neat's-foot oil in its general properties resembles lard oil and is 
used largely for treating leather. 

Sperm Oil (Nondrying). Southern sperm is obtained from the 
head or blubber of the sperm-whale, which is generally found 
in tropical or temperate seas. A large cavity in the head of the 
whale is filled with crystalline matter called " spermaceti." 
Arctic sperm is obtained from the blubber of the bottlenose 
whale, which is found in the northern seas hence the name. 

The crude sperm oil is cooled, so that most of the spermaceti 
separates out, then pressed. The spermaceti is used for making 

Sperm oil has only a slight tendency to oxidize, a low setting 
point, and the lowest viscosity and specific gravity of all fixed 
oils. It is a valuable lubricant for high-speed spindles in textile 
mills, being generally used mixed with low-viscosity mineral 
oils (5 to 25 per cent sperm). 

Whale Oil (Semidrying) . Whale oil is obtained from the 
blubber of the Greenland and other whales. The specific gravity 
of whale oil is much higher than that of sperm oil. Whale oil 
has marked drying properties, but the pale grades are used 
successfully as lubricants when mixed in small proportions (5 to 
10 per cent) with mineral spindle oils for textile purposes or as 
cutting oils. Dark whale oils are lower in quality and cannot 
be used for lubrication but are excellent as tempering or quench- 
ing oils used in the manufacture of tools, guns, case-hardened 
materials, etc. 

Seal oil is similar to whale oil and is obtained from the blubber 
of seals. 

Porpoise Oil, Dolphin Oil, and Melon Oil (Nondrying). These 
oils, which are very similar, are obtained from the soft fat of the 
head and jaw of the porpoise and the dolphin. 

Melon oil is made from a melon-shaped lump of fat in the head 
of the dolphin; the crude oils, obtained in the usual way, are 
chilled and pressed to remove solid fat. These oils are used, 
particularly in the United States, for lubricating watches and 
other delicate mechanisms and command a high price. 

Menhaden, Cod, or Other Fish Oils (Semidrying). Menhaden, 
cod, or other fish oils are obtained by boiling fish in large pans 
with steam; after standing some time the oil rises to the surface 



and can be skimmed off. The color varies according to the 
freshness of the fish and the length of boiling. 

Fish oils are chiefly used in the leather industry, but blown 
cod oil, blown in a manner similar to that used for blown rape oil 
and to similar viscosities, has given fair satisfaction in the manu- 
facture of marine-engine oils. Fish oils have also been used as 
quenching and tempering oils. 















Castor oil 

0.960 to 0.966 


530 to 560 

80 to 90 

176 to 186 

0.1 to 6 


Rape oil 

0.913 toO. 916 

12 to 26 

530 to 560 

96 to 108 

170 to 176 



Ravison rape 

0.918 to 0.922 

108 to 120 

178 to 179 

2 to 6 


Blown rape 

0.960 to 0.985 


Cottonseed oil 

0.921 to 0.926 


560 to 625 

100 to 120 

192 to 195 


Linseed oil 

0.931 to 0.936 

-15 to 10 


170 to 200 

192 to 195 



Olive oil 


20 to 50 

475 to 600 

80 to 90 

185 to 196 

3 to 20 


Coconut olein 

0.925 to 0.930 

40 to 70 


8 to 9 

250 to 260 

2 to 20 


Palm oil 

0.922 to 0.925 

80 to 110 


50 to 56 

196 to 202 

10 to 60 


Peanut oil 

0.918 to 0.925 

27 to 37 

540 to 620 

90 to 102 

187 to 191 



Rosin oil 



25 to 115 

70 to 80 

to 35 



Tallow, beef, or mutton 

0.935 to 0.950 

100 to 125 

550 to 590 

34 to 48 


2 to 10 


Tallow oil 

0.913 to 0.918 

32 to 40 

540 to 600 

55 to 60 


1 to 5 


Lard oil 


32 to 60 

500 to 600 

65 to 75 


3 to 25 


Neat 's-foot oil 

0.914 to 0.917 


470 to 580 

65 to 75 


0.2 to 25 


Sperm oil 

0.878 to 0.882 



80 to 94 

120 to 140 

0.5 to 3 


Whale oil 

0.924 to 0.925 

40 to 50 


110 to 130 

187 to 197 

2 to 10 


Porpoise oil 

0.916 to 0.927 

22 to 48 


Menhaden oil ... 

0.930 to 0.933 

20 to 25 


140 to 170 


3 to 6 


Cod oil, fish oil 

0.921 to 0.928 



145 to 170 


1 to 15 


Wool grease 

0.944 to 0.960 

100 to 130 


15 to 30 


50 to 60 


* Single values only. 

Wool Grease. Wool grease is obtained in the process of wool 
washing; the alkaline scouring liquors containing the wool grease 
are run into settling tanks; the fatty matter accumulating on the 
surface is collected and drained in filter bags. The scouring 
liquors may also be treated with sulphuric acid in conjunction 
with injection of live steam; the acid separates the fatty matter, 



and three distinct layers are formed greasy matter on the top, 
water and soda in the middle, and earthy matter at the bottom. 
The extracted grease is dirty and contains water; the water is 
removed by cold and hot pressing, followed by strong sulphuric 
acid treatment. The wool grease thus prepared is known to the 
trade as " Yorkshire grease" and is used in the manufacture of 
rolling-mill, railway, and colliery greases. 


Saybolt seconds 



Absolute viscos- 
ity, centipoises 

Oil or fat 


















Castor oil. . 











Rape oil 











Tallow, beef, or mutton . . 



Lard oil, neat's-foot oil | 
Olive oil, peanut oil ) ' ' 











Cottonseed oil 











Coconut oil, whale oil. . . . 











Sperm oil 











On the Continent a process of wool cleansing by means of 
solvents (ether or carbon bisulphide) is often employed; the 
solvents are afterward recovered by distillation, and the wool 
grease remains behind. Such wool grease is usually distilled 
with superheated steam and produces wool olein and wool stearin, 
etc. One use of wool olein is in the manufacture of wool oils. 



Semisolid lubricants are lubricants that do not flow at ordinary 
room temperatures. Animal or vegetable fats, such as tallow 
or palm oil, or poor cold-test cylinder stock may be classified as 
semisolid lubricants. Most semisolid lubricants are, however, 
made from mineral oils and saponified fats or fixed oils and may 
be divided into two main groups, i.e., cup greases and solidified 
oils or fats. 

Cup greases are boiled greases and consist of 80 to 90 per cent 
of mineral oil mixed homogeneously with 10 to 20 per cent of 
saponified fat, preferably clarified beef tallow. The tallow is 
mixed with lime water and heated in a steam-jacketed kettle 
(60 to 90 Ib. steam pressure) for 3 to 4 hr. until the base for the 
grease is completely formed. The mineral oil is gradually (5 to 
6 hr.) mixed with the base until the right consistency of the grease 
has been obtained, the mixture being constantly agitated 
mechanically or by compressed air. 

The grease is then run out during the next couple of hours, 
during which the consistency becomes gradually softer owing to 
the agitation, notwithstanding that the speed of the stirrers is 
reduced toward the finish. Some manufacturers run the grease 
out of the boiling kettle into a grinding mill, in which all lumpy 
matter and impurities are reduced, and the grease made of a 
uniform consistency (the more the grease is kneaded the softer 
it becomes). 

Grinding the impurities fine does not, however, remove them; 
it is better to strain the grease when it leaves the kettle. This is 
beat done by forcing the grease, when hot and fluid, under great 
pressure through fine layers of gauze. The gauze retains all the 
impurities, so that the grease is perfectly clean when filled into 
the packages. It is surprising to see the amount of impurities 
that can be retained in this way from grease that one might 
consider practically clean. 



The ideal amount of grease made in one batch is 20 to 25 bbl. 

Cup greases should be free from fillers, such as chalk, china clay, 
gypsum (sulphate of calcium), barytes (sulphate of barium), 
asbestos, talc, wax, etc.; they should be free from uncombined 
lime, gritty impurities, rosiri oil, rosin or resinates, mineral or 
fatty acids, alkalies, or other deleterious impurities; the yield of 
ash should be less than 2 per cent for a medium grease and less 
than 3 per cent for a hard grease; the content of water should be 
less than 2 per cent. 

The melting points of ordinary cup greases range from 75 to 
95C., being higher for the harder consistency greases than for 
the softer greases. 

The consistencies of greases range from very soft to very hard 
and are frequently indicated by numbers, as follows: 

No. 1 No. 2 No. 3 No. 4 No. 5 

Very soft Soft Medium Hard Very hard 

The softer the grease the more oil does it contain. 

The mineral oils used for making cup greases are pale mineral 
oils (pale to give the grease a light color). Red oils might quite 
well be used; the drawback is that they do not give the grease 
such a nice appearance as the pale oils. The viscosities of the 
mineral oils used range from 150 to 1200 sec. at 70F. 

Graphite lubricating grease is cup grease that has been mixed 
with from 5 to 20 per cent of amorphous or flake graphite. 

Cold-neck greases are black lime greases made with black heavy- 
viscosity oils and are used for lubricating "cold" rolling-mill 
necks in steelworks. 

Fiber greases are of a " fibrous" nature but contain no fibers 
of any kind. They are usually made by saponifying a fixed 
oil with caustic potash or caustic soda instead of lime and a little 
water. After saponification the water is boiled out, and the 
mineral oil is worked in. Fiber greases of good quality can be 
melted and cooled again without altering their consistency. 

Some fiber greases have very high melting points, ranging from 
145 to 260C. 

Solidified oils or fats are made in a manner similar to that used 
for cup greases but are made cold and with carbonate of soda or 
caustic soda as the saponifying agent in place of limewater. 
These greases may be made in small quantities, as it is a question 


only of mixing the right proportions of the various ingredients 
together, cold or at fairly low temperature, and stirring the mix- 
ture till it sets. It is obvious that the ingredients cannot be so 
perfectly mixed and combined as with cup greases, which are 
boiled; the result is that certain parts of the grease will often 
contain excess soda, which is detrimental to good lubrication. 

Many so-called soap-thickened oils are a kind of solidified 
oil, various soaps being added to a mineral oil. Sometimes 
special " thickeners" are sold for the purpose of increasing the 
viscosity of mineral oils; e.g., aluminum soap is used, consisting 
of 20 per cent aluminum oleate or palmitate and 80 per cent 
mineral oil, in which the soap is dissolved. Mineral oils thickened 
with aluminum soap have a peculiar nonhomogeneous nature; the 
viscosity is unstable, and the oil is of a slimy nature, forming 
threads when dropping. In contact with water and steam the 
aluminum soap is precipitated and clogs the machinery. 

White greases are usually made from aninal fat and a small 
amount of mineral oil, solidified by soap. The melting points 
are lower than the melting points of cup grease, ranging from 
45 to 70C. 

Certain white greases contain finely pulverized mica and are 
sold under the name of mica greases. 

Railway-wagon Grease. The yellow grease used in the axle 
boxes of railway wagons is usually composed of tallow, palm 
oil, soda soap, and water. 

According to a number of formulas quoted by Archbutt, the 
specifications are approximately as follows: 

Per Cent 

Saponifiable oils 30 to 45, occasionally partly replaced by mineral oil 

Anhydrous soap 10 to 30 

Water. 40 to 60 

Insoluble matter 0.02 to 2.8 

Usually 3 to 5 per cent more water is used in the winter greases 
than in the summer ones. 

A good wagon grease should melt at about 40C. p ithout sepa- 
rating; cup greases are unsuitable for railway wagons, as they 
have too high melting points, and when continuously exposed to 
high temperature in the axle boxes the oil separates out, leaving 
the soap behind. 


Rosin grease is made by stirring together rosin oil, slaked lime, 
and usually black mineral oil or neutral coal-tar oil. 

The rosin acids present in the rosin oil combine with the lime, 
forming a soap, which solidifies the mixture of the various oils. 
Water to the extent of up to 20 per cent is sometimes present 
in rosin greases. 

Rosin greases are used to lubricate rough machinery in col- 
lieries and steelworks. 

Hot-neck greases are very hard greases made from heavy resi- 
dues such as wool pitch, stearine pitch, petroleum pitch, heavy 
asphaltic-base petroleum lubricating oils, thickened with soap 
or rosin grease and containing finely pulverized talc or graphite. 
Hot-neck greases are used for lubricating "hot" rolling-mill necks 
in tinplate works and steelworks. 

Pinion greases are closely related to hot-neck greases; they 
frequently contain pine-tar oil and are very sticky and adhesive. 

Special Greases. Gear grease can be made by mixing a heavy- 
viscosity mineral oil with fiber grease or with paraffin wax. 
Such mixtures are reasonably stable when used in the gearboxes 
of motor cars. 

Solidified oils are not satisfactory as gear greases, nor are those 
cup greases the bases for which have been made from rape oil 
or cottonseed oil. Such greases have too high melting points, 
separate under heat, and the soap that is left cakes "and carbon- 
izes. Cup greases made from a tallow-lime base give reasonable 
satisfaction but are also too high in melting point and inclined 
to cake. 

Yarn grease is a mixture of ordinary cup grease or fiber grease 
and cotton waste or woollen yarn, preferably the latter. The 
strands should not be too long 1% to 2^ in. is a suitable length; 
longer strands get entangled, and it becomes difficult to divide 
the grease when applying it to bearings. 

Black floating grease is made by mixing dark heavy-viscosity 
lubricating oils with powdered talc, in about even proportions; 
this grease is still used in some collieries as a car grease. It is 
low in price and causes great friction and wear, but the bearings 
rarely seize or get scored. 

Petroleum grease is either a petroleum jelly (see page 9) or 
a mixture of petroleum jelly with thin mineral oil; these greases 
have low melting points and little lubricating value, but they 


contain no moisture and are for that reason recommended by 
several makers of small ball and roller bearings. 

Scented Grease. Many greases cup grease, solidified oil, etc. 
particularly when made from rancid fats or fatty oils, are scented 
with oil of citronella or with nitrobenzene to cover up the bad 
odor. Such scenting should be discouraged, as it is difficult to 
know whether a scented grease is of good or bad quality. 


Several kinds of solid materials, such as graphite, talc, soap- 
stone, mica, flowers of sulphur, and white lead, are used for 
lubricating purposes. Some of these solid lubricants, as flake 
graphite or mica, possess a tough, flaky, foliated structure which 
enables them to resist pressure without disintegration. Others, 
such as amorphous graphite or flowers of sulphur, are easily 
crushed into a fine powder when exposed to pressure. 

Again, solid lubricants may be so finely divided as to enable 
them to be suspended in colloidal form in a liquid carrier. The 
colloidal graphite preparations aquadag and oildag, made by 
Acheson's process, are examples of such lubricants, being dif- 
fusions of colloidal graphite in water and oil, respectively. 


Graphite. Graphite is the most important of all solid lubri- 
cants. It is not attacked by acids or alkalies or affected by high 
or low temperatures. 

Graphite is also called " black lead" or " plumbago," but these 
names are slowly going out of use. 

Natural Graphite. The greater part of the world's supplies 
of natural graphite comes from Austria, Ceylon, Italy, Bavaria, 
Madagascar, the United States, Canada, Mexico, Japan, Siberia, 
and England. 

Natural graphite is found in two forms flake graphite and 
amorphous graphite the former is of a tough, flakey structure 
and has a pronounced luster, whereas the amorphous graphite 
has no such luster. 

Natural graphite, as it is obtained from the graphite mines, 
contains some impurities, chiefly silica, alumina, and ferric oxide. 

Most of the natural graphite employed for lubricating purposes 
is of the flake variety. The flake formation is retained even if it 
be ground into a fine powder. It is manufactured in several 
degrees of fineness. 



Flake graphite may be used either dry or in admixture with 
semisolid lubricants. It cannot be used mixed with oil in ordi- 
nary lubricators or lubricating systems, because of its high speci- 
fic gravity (2.2), which causes it to separate out and choke 
lubricators, oil pipes, and oil grooves. 

Artificial Graphite. Amorphous graphite is produced artifi- 
cially by Acheson in the electrical furnace. He is able by his 
process to produce graphite of a soft, unctuous, noncoalescing 
nature and almost chemically pure. 

The varieties produced for lubricating purposes are guaranteed 
to contain 99 per cent of pure carbon but usually contain more. 
In one variety of graphite No. 1340 98 per cent of the graphite 
particles are less than ^33 in. in diameter. From this or similar 
graphite Acheson produces what he calls deflocculated graphite 
by kneading it for a long time with water in the presence of a 
vegetable extract, such as tannic acid. The graphite particles in 
this process disintegrate into particles one thousand times less in 
diameter; in fact, Acheson estimates that each particle of the 
"1340" graphite becomes divided into 700,000 particles, a small- 
ness of size bordering on the molecular. The graphite becomes 
diffused in the water in colloidal form, and each particle, being 
protected by an envelope of organic colloidal matter, remains in 
suspension for an indefinite time in the water. 

The graphite exists in the form of hexagonal tilelike particles 
which dispose themselves with their broad faces to the sliding 
surfaces, the particles on opposing surfaces readily sliding over 
one another with little friction. 

Acheson manufactures the colloidal solution of graphite in 
water in the form of a concentrated paste under the name of 
"aquadag." It may be diluted by the addition of pure water 
to the required strength without the graphite's separating out. 
By a further process the concentrated aquadag is mixed and 
kneaded with mineral lubricating oil until all the water is replaced 
by oil; this product is called "oildag" and may be diluted with 
good-quality neutral mineral oil without any appreciable separa- 
tion of the graphite, without "flocculation," as Acbeson calls it. 

"Glydag" is a concentrated preparation containing 10 per cent 
electric furnace graphite (by weight) colloidally dispersed in 
glycerin, a valuable low-temperature lubricant. For certain 
purposes, a mixture of aquadag and glydag may be preferable. 


In Germany, colloidal solutions of graphite have been produced 
commercially by E. de Haen, similar to aquadag and oildag, the 
corresponding names being hydrosol (corresponding to aquadag) 
and oleosol or kollag (corresponding to oildag). According to 
Holde, 1 in both forms of colloidal graphites there are graphite 
particles of a size from 1 to 6/i, but the majority are submicrons 
less than IM in size (1/x equals 0.001 mm.) which are not easily 
separated out by centrifuging, whereas the larger particles from 
1 to GM are easily separated out in this manner. 

Colloidal solid lubricants may be produced from materials other 
than graphite. It will appear that some successful attempts 
have been made with talc and mica. 

Talc. Talc consists of hydrogen magnesium silicate (H 2 Mg 3 - 
Si4Oi2) and occurs as foliated or scaly compact masses. Its 
specific gravity ranges from 2.6 to 2.8. 

The term steatite is restricted to the compact massive varieties 
of talc. 

Soap stone is an impure form of steatite. 

French chalk is talc or steatite in powder form. 

Talc scales feel greasy or soapy, possess a perfect micaceous 
cleavage, have a pearly to silvery luster, and are flexible but not 
elastic, thus differing from mica. 

Talc is very soft and can readily be scratched with the finger- 
nail; it is selected as No. 1 in Mohs's hardness scale, although the 
harder varieties of talc may have a hardness of 2.5 to 4. The 
color of talc varies from silvery white for the best and softest 
varieties to grayish or greenish for the harder steatite varieties. 

Talc resists acids and alkalies and also heat (no water being 
lost below a red heat) and cold. It is obtained chiefly from the 
United States but is found also in many other countries such as 
England (Cornwall), Bavaria, France, Italy, Austria, and India. 

Mica. The name "mica" is applied to a group of minerals 
characterized by the facility with which they split into thin 
lamina which are flexible and more or less elastic. The hard- 
ness of the micas is between 2 and 3, while their specific gravity 
ranges from 2.7 to 3.1. 

The chemical composition is subject to considerable variations 
in different species broadly speaking, there is a group of potash 

Zeitschrift fiir Elektrochemie 2 %i 6 , 1917. 


micas, generally pale in color; and a group of magnesium or ferric 
magnesia micas, usually dark in color. 

All the micas are complex silicates containing aluminum and 
potassium generally associated with magnesium but rarely with 

Water is always present, and many micas contain fluorine. 

Mica is prepared for the market by splitting the blocks of 
rough mica into plates which are cut into the required patterns 
by means of shears. 

The refuse mica when finely ground forms the material used for 
lubricating purposes. The small particles of mica still retain 
their thin lamellar structure. 

Flowers of Sulphur. Flowers of sulphur is not used much for 
lubricating purposes but is used to some extent for curing hot 
bearings. It is a fine powder consisting of pure sulphur largely 
in the form of minute crystals. The specific gravity is approxi- 
mately 2. 

White Lead. White lead is used to some small extent for curing 
hot bearings. It is an extremely fine powder consisting chemi- 
cally of basic carbonate of lead and generally said to have the 
following formula: 2PbCO 3 -Pb(OH) 2 . 



In the early days, when mineral lubricating oils were nearly all 
made from Pennsylvania or Russian crudes, only a few varieties 
were manufactured, and simple physical and chemical tests 
sufficed to identify the oil. This state of affairs no longer exists; 
lubricating oils are now made from a great variety of crudes, and 
great experience is required to judge the merits of an oil on the 
basis of a laboratory analysis. 

The selection of an oil for certain engines or machinery requires 
many years of experience in comparing and testing different 
lubricants under actual running conditions. Laboratory tests 
and investigations alone are of no avail, as chemists usually 
have no engineering experience; on the other hand, lubricating 
engineers cannot develop their experience and judgment without 
the very best chemical assistance; in fact, it is only by coordi- 
nating field engineering experience with careful laboratory investi- 
gations that it is possible to accumulate the kind of knowledge 
that is required to enable one to give sound recommendations 
as to the grades of lubricants that should be selected for a given 
purpose, as well as the best methods of application and use. 

It is a well-known fact that the vast majority of oil firms 
operate on the principle of getting samples of oils in use, analyzing 
these samples more or less roughly, and then offering oils more 
or less similar in character. As the customer in most cases does 
not trouble much about the quality of the oils, as long as the 
"price is right" and as long as nothing serious happens to his 
machinery, the prevailing standard of lubrication is usually 
exceedingly low. The author, who for many years has been in 
charge of a large staff of lubrication engineers, can testify that 
very few works exist where a lubricating engineer, after a thor- 
ough works inspection, cannot point out means by which great 
economies can be affected from the point of view of saving in 
power (with all its attendant benefits), saving in lubricants, 



greater safety of operation, etc., all due to better lubricants or 
better methods of handling them from the moment they are 
received at the stores till the moment the last drop has been con- 
sumed in the works. 

It should not be necessary for a capable lubricating engineer 
to have samples of the lubricants in use in order to recommend 
the correct grades of his firm's products. His general lubrication 
knowledge of engines and machinery and his observations during 
the inspection ought to be sufficient for that purpose. But if he 
is to give an accurate estimate of the possible saving in power 
or consumption to be obtained by introducing better or more 
suitable lubricants, then an analysis of the lubricants in use and 
of the consumption in all departments is required. 

Speaking generally, in order to satisfy certain lubricating 
requirements, the lubricant 

1. Must possess sufficient viscosity and lubricating power 
oiliness to suit the mechanical conditions and conditions of 
speed, pressure, and temperature. 

Too little oiliness means excessive wear and friction; too high 
a viscosity means loss of power in overcoming unnecessary fluid 

2. Must suit the lubricating system. 

When, for example, the oil pipes are exposed to cold, a lower 
cold-test oil is required than when the oil pipes are not so exposed. 

3. Must be of such a nature that it will not produce deposits 
during use exposed to the influence of air, gas, water, or impurities 
with which the oil may come into more or less intimate contact 
while performing its duty. 

The particular physical and chemical tests needed will depend 
on the class of work for which the oil is to be used and will become 
more apparent from the chapters in this book devoted to particu- 
lar types or sections of engines and machinery. 

In the manufacture of lubricating oils it is of the greatest 
importance that the various grades be kept always as closely as 
possible to certain predetermined standards. Engineers who 
have to do with the practical application of oils fully jij>preciate 
this point. For example, a drop-feed lubricator on a bearing is 
set to give a certain feed of oil which has been found satisfactory; 
a new supply of oil is received of a lower or higher viscosity than 
the former supply; the feed of the lubricator will then be either 


greater, which means oil wasted, or smaller, with the result that 
the bearing may run warm. 

Physical and chemical tests of lubricants are therefore of great 
value to the oil manufacturer for controlling the manufacture 
of lubricating oils during the distillation, refining, and compound- 
ing operations, up to the point when the oil is placed in the stores 
ready for shipment. Physical and chemical tests are also 
extremely valuable for the purpose of identifying an oil or for 
detecting adulterations. 

In the following chapters the author will endeavor to show 
the importance of physical, chemical, and mechanical testing 
methods, but with the exception of one or two, which he feels 
may not be generally known, it is not proposed to describe the 

The author has divided " Testing Lubricants " into two sections, 
viz., "Physical and Chemical Tests " and "Mechanical Means 
of Testing Lubricants/ 7 the latter section dealing briefly with 
friction testing machines and works methods of carrying out 
comparative tests on engines and machinery. 


Physical Tests: 

Density and Specific Gravity. 

Coefficient of Expansion. 

Flash Point and Fire Point. 

Volatility Loss by Evaporation. 


Dilution Test. 

Specific Heat. 

Cold Test, Pour Test, and Cloud Test. 

Melting Point. 

Color and Fluorescence. 

Viscosity of Oils. 

Viscosity of Semisolid Lubricants. 



Surface Tension. 

Chemical Tests: 


Sapohification Value. 
Iodine Value. 
Oxidation and Gumming. 



Carbon Residue. 

Asphalt and Tar. 


Impurities (Dirt, Glue, Water). 


Density and Specific Gravity. The specific gravity of a sub- 
stance is the weight compared with that of an equal volume of 
water as unity. 

In the United States and Great Britain the specific gravity is 
the 60F./60F. value, which means that the specific gravity is 
measured at 60F. as compared with water at 60F. as unity. 

On the Continent the 15C./4C. value is generally used, which 
means that the specific gravity is measured at 15C. and com- 
pared with water at 4C. as unity, this being the temperature 
at which water has its maximum density. 

Density in the c.g.s. system (metric system) means the weight 
of 1 ml. (= cubic centimeter) of a substance as compared with 
the weight of 1 ml. of water at 4C. The specific gravity 
15C./4C. therefore represents in the metric system the density 
of the substance at 15C. The 15C./4C. specific gravity is 
obviously less than the 60F./60F. value, but as the coefficient 
of expansion of water is exceedingly small, the difference in value 
is only slight. 

As indicated in the table (page 24) the specific gravities of the 
various fixed oils do not differ much from one another, whereas 
the specific gravities of mineral oils differ considerably, depending 
not only upon the crude itself but also upon the method of dis- 
tillation and refining. 

For oils made from similar crudes by similar methods the 
specific gravity increases with the viscosity. Speaking generally, 
nonparaffinic-base oils have higher specific gravities than paraffin- 
base oils, the difference for similar- viscosity oils being from 0.020 
to 0.040. Oils treated by acid and cracked oils have higher 
specific gravities than oils treated by filtration and uncracked 
oils, respectively. Coal-tar oils and rosin oils hpve specific 
gravities in the neighborhood of 1.0, coal-tar oils always being 
above 1.0. 

The specific gravity is therefore important, since when coupe d 
with other tests it assists in identifying an oil as coming from a 


certain type of crude, etc. The specific gravity has, however, 
no direct bearing on the lubricating value of a lubricant. 

The specific gravity may be determined by pyknometer, 
hydrometer, or the Westphal balance. The pyknometer method 
(specific-gravity bottle or the Sprengel tube) is applicable to all 
liquids and is the most accurate method for lubricating oils. 
The hydrometer and the Westphal balance are less accurate, 
but both methods are capable of giving sufficiently accurate 
results for commercial purposes and are handier to use than the 
pyknometer, especially the hydrometer. 

The Baum6 gravity is measured by a hydrometer and is much 
used in the United States. The conversion of gravity from 
degrees Baum6 to specific gravity can be carried out according 
to the formula 

Specific gravity = o^ + 13Q 

As 1 1. of water weighs 1 kg., the weight of 1 1. of oil in kilograms 
is expressed by its specific gravity. As 1 imperial gallon weighs 
10 lb., the weight of 1 imperial gallon of oil in pounds is equal 
to ten times its specific gravity. This rule cannot be applied to 
American gallons, 1 American wine gallon equalling % imperial 

The Twaddell gravity scale is sometimes used for liquids 
heavier than water, such as coal-tar products, caustic potash, 
sulphuric acid, and other chemicals. To convert degrees Twad- 
dell to specific gravity use the following formula: 

. . , 1,000 + (5 X degrees Twaddell) 

Specific gravity = * - ^ - - ' 

Coefficient of Expansion. The coefficient of expansion is the 
expansion or contraction per unit volume following a change in 
temperature of 1. 

The coefficient of expansion is the same for all mineral oils of 
the same specific gravity and can be taken near enough for prac- 
tical purposes as being: 1 

1 U. S. Bureau of Standard Technologic Paper 77: Density and Thermal 
Expansion of American Petroleum Oils. 





of expansion 

Per F. 

Per C. 

For gasoline 

0.620 to 0.760 
0.780 to 0.830 
0.850 to 0.970 



For kerosene 

For lubricating oils, including fixed oils 

The density of an oil will vary, with a certain change in tem- 
perature, an amount equal to the coefficient of expansion mul- 
tiplied by the number of degrees that the temperature has 

To know the value of the coefficient of expansion is therefore 
useful for converting the gravity measured at a temperature 
different from the standard temperature (which is 60F. in the 
United Kingdom and United States) to the gravity at the standard 
temperature. It is also useful for measuring the stock of oil in an 
oil-storage tank, as the volume must always be corrected to 
represent volume at a standard temperature. 

In correcting the specific gravity for variation in temperature, 
the correction coefficient is not, as is often assumed, the coefficient 
of expansion but the product of the latter and the specific gravity 
taken at the temperature of the oil. It may be useful to show 
how the true correction is calculated. 

The change in volume due to change of temperature is 
expressed in the fundamental formula 

V T = Feo [1 + C(T - 60)] 
where VT = volume of a certain weight of oil at the temperature 


F 60 = volume of the same weight of oil at 60F. 
C = coefficient of expansion. 

The weight of the oil equals the volume multiplied by the 
specific gravity, so that 

X /Seo ^ VT X ST 


Feo X jSeo 

where S 60 , S T = specific gravities of the oil at 60 and TF., 


We can now rewrite our formula as follows : 

Vt * S = F M [1 + C(T - 60)] 


Seo = S T + S T X C X (T - 60) 

In other words, the specific gravity at 60F. equals the specific 
gravity at TF. plus the product of (1) the difference in tempera- 
ture between T and 60, (2) the coefficient of expansion, and (3) 
the specific gravity at !F F. 

Flash Point and Fire Point. The flash point of an oil is the 
temperature at which the oil gives off sufficient vapors to ignite 
momentarily when exposed to a flame or spark. The oil must 
be heated at a uniform rate and not too rapidly, as that would 
give too low a flash point. 

The open flash point is the flash point determined when heating 
the oil in an open cup. 

The closed flash point is the flash point determined when heating 
the oil in a closed vessel, which rather prevents the vapors 
from escaping, so that the closed flash point is always lower than 
the open flash point. The difference is greater the higher the 
flash point of the oil. 

The fire point of an oil is the temperature at which the oil gives 
off sufficient vapors to ignite and continue to burn when exposed 
to a flame or spark. The test is made with the same apparatus 
as is used for determining the flash point, the oil being heated 
beyond the flash point until the fire point is reached. 

No oil is used for lubricating purposes with an open flash point 
less than 300F. The open flash points of all lubricating oils, 
including fixed oils, range from 300 to 650F. The closed 
flash point of a lubricating oil is recorded only for special oils, 
such as air-compressor oils, and transformer oils. 

The apparatus employed for testing flash and fire points varies 
for different countries. Thermometers are usually standardized 
with the bulb and stem at the same temperature. As the stem 
of the thermometer when determining flash and fire points is not 
exposed to high temperature, the results should be corrected by 
adding to the thermometer readings the following degrees 
Fahrenheit : 


275300 5 42645013 

300325 6 45050016 

325350 7 50055020 

350375 8 55060023 

375 40O 10 60065027 

40042511 65070030 

Pensky-Martens apparatus is more widely used for lubricating 
oils in many countries than any other apparatus. 

The Gray instrument is an adaptation of the Pensky-Martens 
apparatus, and the two instruments give identical readings. 

The Abel instrument is used principally for taking closed flash 
points of spirits and illuminating oils. 

Fixed oils do not evaporate until they begin to decompose, 
whereas mineral oils start to evaporate long before their flash 
points are reached. When fixed oils are heated sufficiently to 
give off vapors, destructive distillation has already begun, and 
it will be seen from the table (page 23) that the flash points of fixed 
oils are much higher than for mineral oils of similar viscosities, 
the open flash points ranging from 460 to 630F. 

The difference between the open flash point and the fire point 
of lubricating oils is approximately as follows: 

Difference between Open Flash 

Point and Fire Point, 

Degrees Fahrenheit 

1. Straight mineral distilled lubricating oils 40 to 55 

2. Cylinder oils (undistilled) 50 to 75 

3. Mixtures of mineral distilled oils with 

cylinder oils or fixed oils 40 to 75 

The evaporation point is the temperature at which an oil begins 
to give off vapors; this temperature is normally about 150 to 
180F. lower than the flash point but is so difficult to determine 
accurately and depends so much on the human element that its 
determination is of no practical importance. 

Volatility Loss by Evaporation. Oil exposed to a high tem- 
perature for a certain number of hours loses a certain amount in 
weight, which is called "loss by evaporation/' 

The oil is usually heated in an open beaker, and experience 
shows that the loss by evaporation is greatly influenced by the 
size and shape of the beaker, the amount of oil used, air currents, 
etc. When giving figures for loss by evaporation, one should 


therefore state all such particulars, for the test to be of any 
value at all. 1 

The evaporation test is seldom of any great value; if a lubri- 
cating oil has an open flash point above 300F., the loss by evapo- 
ration will usually be of no importance; the flash point will 
prove a safe guide as to whether the oil contains light petroleum 
fractions of a kerosene or gasoline nature. 

Where oils are known to be contaminated with low-flash prod- 
ucts, the evaporation test can be used to determine the percent- 
age present of these products, e.g., with used oils from the 
crankcase of gasoline or oil engines. 

Lubricating oils used for high-vacuum pumps (in the manufac- 
ture of electric bulbs) must have a low volatility in vacuum and 
should have their vapor tension tested, when exposed to vacuum 
and at a temperature approximating the working temperature. 

Transformer oils are often subjected to the evaporation test, 
as many of these oils are low-flash oils (occasionally flashing 
below 300F.), and the loss by evaporation during use may easily 
become an important feature. 

Air-compressor oils are sometimes tested with advantage for 
evaporation losses, particularly when the compressed air is used 
for tunnel work or for operating tools or engines in confined 
spaces or in underground mines, as the presence of an appreciable 
amount of oil vapor in the compressed air will affect the eyes and 
throats of the workers. 

Archbutt has designed a simple vaporimeter, 2 in which the oil 
is placed in a boat inside a %-in. internal-diameter tube, through 
which is passed hot air or steam, the whole heated to the desired 
temperature by means of a gas burner. The apparatus appears 
to give very consistent results and to lend itself well to standard- 

The vessels used for evaporation tests should be porcelain or 
glass; metal vessels have a catalytic effect, which influences the 

From tests carried out by Archbutt and others, it is evident 
that no simple relation (if any relation at all) exists between the 
volatility of an oil and its flash point. 

1 The evaporation per square centimeter of surface is a better guide than 
the loss per gram of oil! 

2 ARCHBUTT and DEELBY, "Lubrication and Lubricants," p. 215. 


Distillation. In rare cases lubricating oils are subjected to dis- 
tillation tests with a view to finding out the percentages of low- 
viscosity, medium- viscosity, and high- viscosity oils of which they 
are composed. 

All lubricating oils are mixtures of hydrocarbons having differ- 
ent viscosities; and while it is not of any considerable interest 
further to analyze from a distillation point of view the main 
types of oils referred to on page 10, yet it may be of interest to 
find out whether a certain lubricating oil is a mixture of cylinder 
stock and a lower viscosity distilled lubricating oil and, in that 
case, what the percentage of cylinder stock amounts to. 

No standard method has been adopted for a distillation test 
of lubricating oils, nor does this test seem to be of particular 
interest to ordinary consumers. On the other hand, it may be of 
considerable interest to oil refineries or lubricating-oil companies 
with a view to finding out the characteristics and component parts 
of competitive products. 

Dilution Test. This test is now one of the A.S.T.M. stand- 
ards by which is determined the amount of gasoline that has 
diluted the crankcase oil in automobile engines (see page 507). 

Specific Heat. The specific heat of a lubricating oil means 
the amount of heat required to raise the temperature of 1 Ib. of 
oil 1F. or 1 kilo of oil 1C., as compared with the amount of heat 
required to heat 1 Ib. of water 1F. or 1 kilo of water 1C., respec- 
tively. The specific heat of water is therefore 1.00. 

A considerable amount of work in connection with specific 
heats of oils has been done by Prof. Charles F. Mabery. 1 

Professor Mabery has shown that the specific heats of the 
paraffin series of hydrocarbons are higher than the specific heats 
of the naphthenes, olefins, and other hydrocarbons less rich in 
hydrogen than the paraffins. 

The specific heat is higher for the lower viscosity oils than for 
those of higher viscosity, although the difference amounts to only 
a few per cent. The specific heat also increases slightly with 
increasing temperatures. 

For practical purposes, however, the specific heat mpry be taken 
as follows: 

1 Proceedings American Academy of Arts & Sciences, vol. 37, p. 20, March, 



Mineral lubricating oils 

of hydro- 

heat at 

Paraffin-base distilled low- viscosity oils 

v> n -H 2n 4- 2 


Russian oils and heavy-viscosity Pennsylvanian 
oils, etc 

^ / n*l 2n 


Many asphaltic-base oils 

OnH^n 2 


The preceding values for specific heat show a characteristic 
difference between the different lubricating oils, which is of some 
importance in connection with lubrication, as the frictional heat 
developed during the operation of machinery heats the oil film, 
thus reducing its viscosity. The lower the specific heat the 
greater will be the temperature rise of the oil in the film, and there- 
fore the greater will be the reduction of viscosity. 

Setting Point or Cold Test, Pour Test, and Cloud Test. 
When lubricating oils are cooled they do not congeal suddenly, 
as, for example, water congeals when it turns into ice, but, being 
mixtures of products of different nature, they gradually become 
more and more viscous until they finally set solid; the tempera- 
ture at which they congeal is called the " setting point, 77 or "cold 
test. 77 

The lowest temperature at which the oil will flow or pour out of 
a receptacle is usually taken as being 5F. above the setting 
point and is called the "pour test. 77 

The temperature at which the oil starts to become cloudy 
paraffin wax separating out is called the "cloud test, 77 but it is 
difficult to determine this temperature with accuracy, and the 
cloud test is nowadays rarely spoken of in connection with 
lubricating oils. 

Stirring. When the setting point of mineral oils is being 
determined, the oil must not be stirred, as by stirring the network 
of solid hydrocarbons is broken up, and the setting point will 
be from 5 to 10F. lower than when the oil is cooled without 
stirring. Archbutt, however, recommends stirring when testing 
fixed oils. 

Russian oils, Californian, and other nonparaffinic-base oils 
have no cloud test, as they contain no solid paraffin; their setting 
points are therefore lower from 20 to 40F. lower than those of 
paraffin-base oils. 


Sometimes heavy-viscosity paraffin-base oils become chilled 
during transit in cold weather; as a result, the amorphous wax 
begins to solidify in oily lumps throughout the body of the 
oil. The oil will therefore be much thicker than its standard 
(real) viscosity and will not be homogeneous. To bring the 
oil back to its normal viscosity, it is necessary to heat it suffi- 
ciently to melt the paraffin wax, say, to 160 or 170F., the melt- 
ing points of paraffin wax ranging from 100 to 130F. 

When light-viscosity lubricating oils become chilled, some of 
the paraffin wax sometimes crystallizes out in the form of shiny 
needles floating in the oil; they will dissolve in the oil only 
if heated to a temperature above their melting point. 

Heating. When testing the setting point of oils containing 
paraffin wax, they should, for the reasons just given, always be 
previously heated to a temperature of 160 to 170F. 

Cooling. The oil should be cooled slowly, as rapid cooling 
means that the setting point, as determined, will be too low. 
This is particularly important for fixed oils. The test tube 
or bottle containing the oil should therefore be placed inside 
another tube ^ in. larger in diameter, the air space between 
the tubes preventing too rapid cooling. 

Apparatus. There is a variety of apparatus employed for 
testing the setting point, pour test, and cloud test of oils. Some 
aim at determining the setting point as the temperature at which 
the oil in the vicinity of the thermometer ceases to flow when the 
vessel or tube is tilted for 10 min. Others aim at determin- 
ing the pour test as the temperature at which a definite quantity 
of the oil will just flow from end to end of a test tube of definite 
dimensions when placed horizontally or inverted; this method 
is used largely for cylinder oils and black oils. 

The determinations of setting point or pour test are usually 
accurate within 5F. 

Cooling Mixtures. For oils congealing above 35F., pounded 
ice is used. For oils congealing at from 35 to 5F., a mixture 
of snow or pounded ice and salt is used, the salt preferably being 
added gradually to bring the temperature down 5F.*at a time. 

For oils congealing below zero, solid carbon dioxide can be 
used or calcium chloride (crystals) and ice ( 40F.), or solid 
carbon dioxide may be added to dry acetone, by which a tem- 
perature as low as 70F. can be obtained. 


The setting point of an oil must be low enough so that the 
oil will flow readily under working conditions and so that a suffi- 
cient amount will reach the bearings or parts to be lubricated. 
Many mishaps have been caused by the oils solidifying in the 
lubricators or refusing to run through exposed oil pipes to the 
bearings. While, therefore, in tropical or warm climates the set- 
ting point of lubricating oils ordinarily is of no importance, 
this feature is certainly important in temperate climates and 
particularly in colder climates like those of Canada, northern 
Scandinavia, and northern Russia. 

Oils used for refrigerating machines and other special machines 
must always have low setting points, independent of the climatic 
conditions. Low setting point must be given special considera- 
tion in connection with engines or machinery operating in the 
open, such as railway rolling stock, certain machinery in mines 
and quarries, airplanes, automobiles, etc. Oils used for engines 
or machinery operating inside buildings do not require the same 
consideration as regards setting point. 

Whenever low-setting-point oils are required, the winter con- 
ditions are, of course, more severe than summer conditions; 
and so, frequently, two sets of oils are used for summer and for 
winter use, respectively. 

Melting Point. The melting point of an oil, fluid at ordinary 
temperatures, is the same as its setting point, or rather its pour 
test; in fact, the latter is sometimes determined by freezing the 
oil solid, then allowing it to melt exposed to the room tempera- 
ture, under continuous stirring, until the oil starts to pour. The 
melting point of fats, lubricating greases, or oils nonfluid at 
ordinary temperatures is not a definite temperature, as they 
become soft and gradually melt, when heated. 

Melting points of fats and greases may be determined in 
several ways, as described by Archbutt; 1 no uniform system 
has been agreed upon, and there are great discrepancies between 
the results when different apparatus is used and by different 
observers. 2 

1 Op. cit., pp. 223-229. 

2 Very low-melting-point greases (m.p. 20 to 30C.) are required for 
gearboxes of automobiles, pneumatic tools, etc. Such greases may simply 
be poor cold-test cylinder stock. 


Low-melting-point greases (m.p. 40 to 50C.) are required 
for railway axle-box lubrication. Medium-melting-point greases, 
such as cup greases or solidified oils (m.p. 75 to 95C.) are used 
for general lubrication. High-melting-point greases are used 
for high-temperature bearings for rotary cement kilns, dryer 
journals and calendar journals on paper machines, beater bear- 
ings, etc. (m.p. 150 to 250C.). 

Color and Fluorescence. Fixed oils are transparent and either 
almost colorless or slightly yellow or greenish yellow in trans- 
mitted light. 

Distilled mineral oils are transparent and range in color from 
water-white, through yellow, to deepest red in transmitted light. 

Undistilled mineral oils cylinder stocks are very dark in 
color; dark cylinder stocks range from dark brownish red to 
black, whereas the filtered cylinder stocks are lighter in color 
and range from deep red to deepest red in transmitted light. 

Lovibond's tintometer may be used to determine by com- 
parison with standard colors the color of oil in, say, a 1-in. cell. 
The darker the oil the higher is the color number. If determined 
in a cell of different thickness, the thickness of the cell should be 
stated, so that the color number may be calculated in terms of 
a 1-in. cell. 

An apparatus has been designed employing a photocell. The 
oil is placed in a glass container, and the photocell measures 
the amount of light absorbed in passing through the oil. By 
interposing glasses of different colors between the source of light 
and the cell, the absorption of the various colors can be meas- 
ured, thus giving a complete picture of the true color of the oil 
expressed in shades of yellow, blue, and red. 

Wearham has designed an apparatus that gives the color of 
an oil or any other material, liquid or solid, in terms of per- 
centages of primary red, blue, and yellow. The color determined 
in this manner is definitely expressed and can be reproduced 
in the apparatus to act as a standard for matching colors at the 
refineries or oil-blending works. 

The color of an oil in reflected light is called "bloom" or 

Paraffin-base oils have a greenish bloom; Russian and many 
asphaltic-base oils have a bluish bloom. Dark cylinder stocks 


are dark brown or black, whereas highly filtered cylinder stocks 
show the fluorescence clearly and are usually green, being pro- 
duced from paraffin-base crudes. 

Oils that during use have been oxidized (turbine oils, crank- 
case oils, etc.) almost immediately change their bloom and assume 
a brownish color in reflected light. Oils that contain mois- 
ture become cloudy or even opaque and appear to be darker in 
color than when dry. 

Coloring matter in oil consists of very complex unsaturated 
hydrocarbons, which easily decompose under heat or when 
exposed to oxidation. Dark-red oils when used for internal- 
combustion engines or air compressors therefore produce more 
carbon than pale oils, and dark-colored circulation oils are 
more inclined to produce deposits in steam turbines and enclosed- 
type steam engines than are pale oils. Similarly, dark cylinder 
oils produce more carbon than filtered cylinder oils when used for 
steam engines employing superheated steam. 

Where oils are not exposed to great heat or oxidation, it is 
immaterial whether they are lighter or darker in color. 

Viscosity of Oils. The viscosity of an oil is a measure of its 
resistance to flow its internal friction and is inversely propor- 
tional to its fluidity. A viscous or high-viscosity oil is "thick" 
and flows with difficulty; a low-viscosity oil is "thin" and flows 

The most accurate method of determining viscosity which is 
used chiefly in science is that of Poiseuille, by measuring the 
rate of flow of the oil through a capillary tube (a very narrow 
tube) under known conditions of temperature and pressure. 

For commercial purposes the viscosity is usually determined by 
measuring the time taken in seconds for a certain volume of oil 
to flow out through a short vertical tube of standard dimensions. 

Poiseuille's Method. Poiseuille's method is based upon two 
facts, which he proved experimentally: 

1. That the rate of flow of liquid through a capillary tube of 
suitable dimensions is proportional to the pressure and inversely 
proportional to the length of the tube. 

2. That the rate of flow through a capillary tube of cylindrical 
bore is proportional to the fourth power of the radius of the bore. 

The viscosity of the fluid in absolute measure is then given 
approximately by: 


Absolute viscosity = 17 = ^ 

where g = acceleration due to gravity. 
d = density of the liquid. 
h = mean head. 

r = radius of the bore of the capillary tube in centimeters. 
t = time in seconds. 

v = volume of liquid discharged in cubic centimeters. 
a = length of the tube in centimeters. 
Reynolds has shown that Poiseuille's formula is substantially 

correct when is less than 700 where v is the mean velocity of 

the fluid. 

To obtain the true viscosity, corrections must be made 

1. For the viscous resistance to the flow of the liquid at the 
ends of the tube. 

2. For the abnormal flow of the liquid on first entering the tube. 

3. For the kinetic energy with which the liquid leaves the tube. 

4. For the resistance due to surface-tension effects at the dis- 
charge orifice. 

Corrections 1 and 2 have not yet been devised, but errors 
due to these effects may be reduced to very small proportions 
by making the tube long. Correction 3 is made by deducting 
from the mean head a quantity v*/g; this correction may be made 
very small by using a tube so narrow and so long that the move- 
ment of the liquid is very slow. The error due to surface-tension 
effects, which may be serious, is so variable that correction 4 is 
best eliminated altogether by immersing the discharge orifice 
in the liquid and making a suitable deduction from the head. 

The absolute viscosity of a liquid may be defined as the 
force that will move a unit area of plane surface with unit speed 
relative to another plane surface from which it is separated by a 
layer of the liquid of unit thickness. 

The absolute viscosity is therefore correctly expressed in dyne 
seconds per square centimeter but is usually referred to as dynes 
per square centimeter. 

The term " poise " has been coined from the name of Poiseuille 
and signifies 1 dyne sec. per square centimeter. As an absolute 
viscosity of 1 poise means a fairly viscous oil, the term "centi- 



poise" is used, representing 0.01 poise, just as 1 cm. equals 

0.01 m. 

The kinematic viscosity in the c.g.s. system is obtained by 

dividing the absolute viscosity with the specific gravity of the 
oil. These kinematic viscosities are called 
" stoke" and "centistoke," corresponding to 
poise and centipoise. 

Ostwald's viscometer (Fig. la) is capable of 
giving the true relative viscosity in terms of 
the viscosity of water or other standard liquid, 
which is called the specific viscosity, when com- 
pared with the viscosity of water at 20C. as 
unit. It consists of a glass U tube, one limb of 
which is a capillary tube from a to b. A known 
volume of oil is introduced into the wide limb 
at c and sucked up at d, until the level is 
above e. The time occupied in flowing back 
through the capillary ab while the level falls 
from e to a is noted. 

If id and i\d\ are the time of flow and density 
of the liquid and water, respectively, and if the 
viscosity of water is taken as unity, then the 

FIG. la. Ostwald's 


Specific viscosity == 7-7- 


As the tendency is more and more to indicate viscosities in 
absolute units, there is a growing need to have a commercial 
apparatus for this purpose. Recently, such apparatus have 
been proposed, one being a modified Ostwald viscometer and the 
other the Ubbelohde suspended-level method, both of which have 
been published by Committee D-2 of the A.S.T.M. 

Commercial Viscometers. The most widely used instruments 
are Say bolt (United States), Redwood (Great Britain), and 
Engler (Continent). 

These instruments are described in greater detail in most 
standard handbooks. 

The viscosities for Saybolt and Redwood are given in seconds, 
whereas with the Engler instrument, the Engler number equals 
the outflow in seconds divided by the efflux time of water at 20C., 
which is 50 to 52 sec., varying slightly for different instruments. 



The Engler number is therefore, a kind of specific viscosity and 
can be converted into absolute viscosity by the chart on page 54. 

All three viscometers have efflux tubes so large and compara- 
tively short that a large proportion of the energy of flow, particu- 
larly for low- viscosity liquids, is carried 
away in the issuing stream of liquid and 
not used for overcoming the fluid 
friction inside the efflux tube. For 
example, with the Engler viscometer the 
percentage of energy used in overcoming 
fluid resistance within the efflux tube is 
95 per cent or more in the case of a 
heavy- viscosity oil, whereas with water 
it is only about 12 per cent. 

The relation between the absolute 
viscosity of an oil and its kinematic 
viscosity, as measured by Saybolt, Red- 
wood, or Engler viscometers, is fairly 
uniform for medium- or heavy-viscosity 
oils, but when the kinematic viscosity is 
lower than, say, 100 sec. Saybolt, the 
absolute viscosity falls away more 
rapidly than the corresponding kine- 
matic viscosity figures. 

Viscometers should be calibrated by 
testing the viscosity of a standard liquid, 
as water, mixtures of glycerin and 
water, cane-sugar solutions, etc. Rape 
oil or sperm oil as recommended by 
some for standardizing viscometers FlG * lb - 

varies too much in viscosity to be used for standardizing purposes. 

A commercial apparatus for determining the kinematic vis- 
cosity in the c.g.s. system is proposed by Raaschou (Laboratory 
of General Technical Chemistry of the Technical University of 
Denmark). This viscometer permits viscosity determinations 
over a wide range. It is designed with a view to reducing to a 
minimum deviations originating from the difference in the capil- 
lary action of liquids. 

The viscometer whose dimensions are given in Fig. 16 is 
U-shaped, One side is a wide glass tube. The other side, the 


capillary side, is by the bulb portion A with a capacity of about 
2.5 ml. divided into an upper and a lower capillary tube. 

The wide tube has an annular red mark a at the bottom and 
millimeter division downward as shown. A thermometer is 
placed axially in the tube sliding in a cork stopper with lateral 
slit. The thermometer serves partly to control the temperature 
of the sample, wherefore it should be calibrated (standard ther- 
mometer), and partly to adjust the heights of liquid before the 
measurement commences. 

On the capillary-tube side, the lower capillary tube has a white 
mark I, and two blue marks II and III. Just below and above 
the bulb A, two red marks IV and V are placed. 

It is practical to place the viscometer in a metal stand, when 
in use, whereby it is protected against breakage, and the capillary- 
tube side is connected to a suction device, e.g., a syringe. For 
the connection a rubber hose with inside diameter 4 mm. and out- 
side diameter 7 mm. may conveniently be used. Near the suc- 
tion device a three-way cock is inserted to make a connection 
between the viscometer and the atmosphere and also between 
the viscometer and the suction device. The rubber hose is given 
such a length that the three-way cock and the suction device 
may rest on the work table. 

When the viscosity determination is to be made, the stand with 
the viscometer is placed in a thermostat in which the temperature 
can be adjusted with an accuracy of at least 0.1C. 

Procedure. The viscometer may be employed in two ways, 
A and B, according to whether the oil sample is thin (v from 
about 1.5 to about 100 cS), or viscous (v from about 100 to 
5,000 cS). 

A. The viscosity of liquids of comparatively low viscosity is 
determined by introducing the sample into the clean viscometer 
until the liquid surface in the wide tube is from 3 to 5 mm. below 
the mark a. Then the stopper with the thermometer is placed 
in the wide tube, and the viscometer is immersed vertically into 
a liquid bath of the temperature at which the viscosity is to be 
determined so that the mark V on the capillary tube side is a 
little below the surface of the bath. 

When the sample has assumed the temperature of the bath, the 
liquid surface at the wall in the wide tube is adjusted to the mark 
a by pushing the thermometer. Then the sample is sucked up 


into the capillary-tube side until the liquid surface is a couple of 
millimeters above the mark V, when the three-way cock is opened 
to the atmosphere and the efflux time of the liquid from the mark V 
to the mark IV (the red marks) is determined by means of a stop 

A mean figure t is taken for the efflux time in seconds of at least 
two determinations which do not differ by more than 0.5 per cent 
from each other. 

B. The viscosity of liquids with relatively high viscosity is 
determined by introducing the sample into the viscometer as 
described above, and causing it to assume the test temperature. 
Then the liquid surface is adjusted by pushing the thermometer so 
that the miniscus, when the liquid column falls, will settle at the 
mark I on the capillary-tube side. The liquid is then sucked up 
into the lower capillary tube until the surface is a couple of milli- 
meters above the mark III, the three-way cock is opened to the 
atmosphere, and the efflux time from the mark III to the mark II 
(the blue marks) is determined by means of a stop watch. 

A mean figure t is taken for the efflux time in seconds of at least 
three determinations which do not differ by more than 1 per cent 
from each other. 

The kinematic viscosity determined in this manner is then : 

v = k A - t (method A) 

v = k B t (method B) 

where kA and ko are calibration constants which have been deter- 
mined according to the above method with a liquid of known 
kinematic viscosity, viz., a mineral oil having a v of about 150 cS 
at 20C., and whose viscosity is accurately determined by The 
National Physical Laboratory, Teddington, England. 

When particulars of the viscosity of an oil are given, important 
details are often omitted. Sometimes the name of the viscometer 
is not given or the temperature at which the viscosity is taken. 
Furthermore, if it is desired to calculate the absolute or the 
specific viscosity, it is necessary also to know the specific gravity. 

The National Petroleum News has published an excellent chart 
(see page 54) by means of which kinematic viscosities may be 




"T ~~ 






r 1,000000 




: 1 


r 20,000,000 


[ 10,000,000 q 

2W<W J 







r 30(^000 









r5A)PO 2Wj 

500,000 ~ 




r^QOOO l(W0 2 



r^ 00 ^ 00 


400,000 -: 

r2pOQ000 J 

2001000 H 

1 5(1000 


W o, 



io<p j 


r I,000j300 

woqpoo d 

: 20QDOOJ 

r 20,000 




fsoopoo JOQOOO _ 



L soqpoo 

^J 500,000 ~ 

tf) - 

-30CPOO H 

w (0 3QOOO- 

r 10,000 


x 1 ho w S 300POO - 


L 20 q 000 -S ^^^ 

? 75 : 

.5 C 2 QOOO^ 


-2oqpoa c 

| W 1[ 50 ^ 2fl(V)a) ^ 


; C 3QOOO" 


~ 5.000 ^ 


i- Qg-^ : ^ <U : 

r o 

Q_ ~ 

- i) 

: <D 

o r*?w o ~ 

(0 : 

y 20,000 

.- QJ : 

: Q) 

- 1 rtnfvin W 

i 1-^0 ' " 

*- OftfVM " 

1 IOQOOO * - 

rJ^OO v. 


'" Q7-^ ^~70 a: ^ IOOPOO ~- 

/WOO ~ 

r /) ' 


: g> 


=_nn . (fl 3 

O : 

r lO^OOOd 

w "o ^ 

U L ~ 

: Q 

:BO/XK) ^ 

**!>** y^l 

>. 2 

: iy -J 

* t 1 

b 1,000 i 

i-jqooo ^ 

c " 

tf) ~ 

. - iOOO H 

^ 'o ^^ 

- 0^ 

: _ O 

-J JH ; 

O i$00 * 

O -: 

r?ODOO .5 2000 -i 

- 2000 -^ 

r cftn c 

r 20,000 o 

20,000 -i 

; J : 

U ^ 'T ; 

7500 UJ 

- ^ 

*" : 

> $00-: 

; 1 

.!? o -i 

: TJ 

~Q "" 

:ioj)oo ^ ^B 

> i/i : 



r!0,000 Q) 

r tt^ 

" tqwod 

^ J3 -i 

s i 


HMD 1 - 

0) ^ 

2 500- 



tyjoo i 

c ^ 

i-jpoo 2M-i 

I 300-; 



po J 

500 -| 

fW 100: 

< 200" 




300 -^ 

r -. 



200 i 

: 1,000 so- 


I" 30 

; 1,000 

1,000 i 

100 - 

-SOO 20 ^ 




-300 :(J 

3Q ] 



300 -: 


r200 ^ 

26 - : 


200 -i 

L 5-i 


so 4 

* j 


20 -j 

HOO : 


100 r 












r '-^ 


5 j 






37 * 

57 ^ 

Viscosity conversion chart. (National Petroleum News.) 


readily converted from the values obtained by one instrument 
to those obtained by another (Saybolt, Redwood, Engler, and 
kinematic c.g.s.) simply by drawing a horizontal line. 

Connecting the point marking the kinematic viscosity in 
centistokes on the vertical line to the left with the point marking 
the specific gravity on the vertical line to the right, this line crosses 
the vertical line for absolute viscosity (second from the left) at a 
point giving the absolute viscosity in centipoises. 

Temperatures. Viscosity figures quoted at 70F. are going out 
of use, and rightly so, as the oil is very rarely at this temperature 
during actual use. For oils used externally, the important vis- 
cosities are those taken at 104F. (40C.) and 140F. (60C.), 
as the temperature of the oil film will range somewhere between 
30 and 50C. for ordinary bearings and between 50 and 70C. 
for bearings in steam turbines, enclosed high-speed steam engines, 
and many internal-combustion engines. 

For steam-cylinder oils the viscosity is usually taken at 212F. 
To test the viscosity at higher temperatures, in addition to the 
212F. figure, does not appear to be of any value; although vis- 
cosity is important, there are other and more important properties 
than viscosity which determine the lubricating quality of cylinder 

For oils like air-compressor oils and internal-combustion engine 
oils, where fairly low-viscosity oils are used, the viscosity should 
be measured at 104 and 212F. 

When the setting point of an oil is known and at least two vis- 
cosities, preferably three, a viscosity curve can easily be drawn 
from which can be measured the approximate viscosity figures 
at intermediate temperatures. Commercial measurements of 
viscosity are usually accurate within 1 per cent. 

Oils change in viscosity with change in temperature (ranging 
normally from 0.6 to 6.0 per cent per degree Fahrenheit between 
104 and 212F.). The change per degree Fahrenheit is greater 
for mineral oils than for fixed oils, greater for high-viscosity than 
for low- viscosity oils, greater (at high temperatures) for asphaltic- 
and naphthenic-base than for paraffin-base oils. The increase per 
degree Fahrenheit becomes very great when approaching the 
setting-point temperature of the oil. As nonparaffinic-base oils 
have low setting points, their increase in viscosity at low tem- 
peratures is usually less than for paraffin-base oils. Hence, for 



cold conditions or climates the former oils have better viscosity 
curves at the working temperatures. 

The following figures show the difference in character between 
paraffin-base and nonparaffinic-base oils. 

fornian oil 


base oil 



Saybolt viscosity 

At 212F 






At 140F 






At 104F 






Cold test. . . 






Average change in 

viscosity per 

degree Fahren- 


From 21 2 to 140F. 






From 140tol04F. 






At temperatures below 104F. the viscosity curves are very 
steep, and the increase in viscosity per degree Fahrenheit is still 
higher, particularly so for oils having poor cold tests, as is the 
case with paraffin-base oils, or mixtures containing poor cold-test 
cylinder stock or a large amount of fixed oil or fat. When the 
viscosity curve is very steep around 70F., it means that if the 
oil is supplied through gravity feed oilers, siphon oilers, and 
the like, even slight changes in temperature will appreciably 
affect the oil feed an undesirable feature. 

At high temperature the fixed oils maintain their viscosities 
remarkably well, which is one of the minor reasons why fixed 
oils are better lubricants for bearings inclined to heat because of 
bad mechanical conditions, excessive bearing pressures, etc. 

Variations in viscosity due to temperature changes are in 
Europe often expressed by formulas that give the steepness of the 
viscosity curve on the basis of absolute viscosities ; it is unfortun- 
ately cumbersome to apply these formulas. In the United States 
the so-called " viscosity index " is often employed, expressing the 
variation in viscosity in relation to two oils chosen at .random 
(Pennsylvania oil, 100; asphaltic oil, 0). The disadvantage 
is that owing to modern improved methods of oil refining, we now 






Absolute viscosity 




centi poises 













c . 







































































































































2 5 

































3 3 












17 1 


n oils 

( 700 
I 1,400 




10 5 

* The viscosity figures given in the foregoing table are only approximately correct, but 
they will serve as a guide to convert viscosities from one "viscosity language" into another. 
Furthermore, this table will help the reader to determine the nearest grade of oil correspond- 
ing to those oils which the author recommends throughout this book, the viscosities of 
which will be referred to only as one of the numbers from 1 to 18, indicated above. 

produce oils that may get placed outside the interval to 100; 
we get oils with a viscosity index above 100 and also oils with a 
negative viscosity index. The ideal would be to find formulas 
that in a simple, practical manner would express the viscosity of 
an oil in absolute units at different temperatures, based upon the 
viscosity of that oil at two temperatures, its specific gravity, 
and/or its molecular weight. 

The viscosity of oils when measured under great pressure 1 is 
much greater than the ordinary viscosities, which are measured 
under atmospheric pressure. Mineral oils vary considerably 
among themselves and generally have a high rate of increase in 
viscosity with pressure. The fixed oils, however, do not greatly 
differ from each other and show an increase in viscosity at 1,000 
atmospheres to approximately four times their values at atmos- 

1 Experiments by Dr. T. E. Stanton and Mr. J. H. Hyde, National 
Physical Laboratory, Teddington. 


pheric pressure, compared with anything between ten and 
twenty-five times the corresponding values for the mineral oils. 

When oils are standardized for viscosity, it is customary to 
allow a manufacturer's variation of 4 to 6 per cent above and 
2 to 3 per cent below the standard; the permissible variation 
above the standard is the greater of the two, because too high a 
viscosity is usually less objectionable under the conditions of 
service than too low a viscosity. 

Comparing two oils having the same kinematic viscosity, say 
the same Saybolt viscosity, the heavier oil of the two (the one 
having the highest specific gravity) really is the more viscous, 
because, being heavier, it flows out of the viscometer more rapidly 
than it would if it were lighter in specific gravity. The real vis- 
cosities of nonparaffin-base oils are therefore about 5 per cent 
higher than their kinematic viscosities lead one to expect when 
comparing them with paraffin-base oils. Much confusion has 
been caused by expressing the viscosities of oils in terms of kine- 
matic viscosities, and it would be very desirable if users of oil 
would insist upon viscosities' being measured or specified as 
absolute viscosities, which represent their true viscosities and con- 
sequently form a much better basis for comparing the suitability 
of different oils. 

The viscosity is frequently the most important property of a 
lubricating oil. With perfectly lubricated frictional surfaces 
(complete oil film), the friction is directly proportional to the vis- 
cosity of the oil, and mineral oils are always used. With less per- 
fectly lubricated surfaces the oiliness of the oil becomes important, 
and compounded oils are frequently used. Under very severe 
conditions of pressure (incomplete oil film) the viscosity value 
of the lubricant is no guide at all; the oiliness becomes the all- 
governing factor, and fixed oils or oils rich in fixed oils have 
to be used. 

Speaking generally, high-viscosity oils are required for condi- 
tions of high temperature or great pressure or slow speed. Low- 
viscosity oils are required for conditions of low pressure or high 
speed. For low-temperature conditions it is low setting point 
that governs the selection of oil more than viscosity. 

When blending oils of different viscosities in certain propor- 
tions, it is important for oil manufacturers to be able to find out 
the viscosity of the blended oil. The A.S.T.M. has made an 


excellent chart from which, by merely drawing one line upon it, 
the viscosity of the blended oil can be obtained with great 

Viscosity of Semisolid Lubricants. Cup greases, solidified 
oils, etc., are generally sold in five standard consistencies very 
soft, soft, medium, hard, and very hard. 

The consistency varies with the temperature of the grease, and 
the consistency of a sample of grease is a matter of personal 
judgment developed by experience. Several attempts have been 
made to devise an instrument for measuring the consistency 
viscosity of a grease, but none has been of any practical value. 
Several instruments have a weighted needle, which is allowed to 
penetrate for a certain number of seconds or until it stops because 
of skin friction; the observations are very irregular even with the 
same sample of grease, owing to local variations in consistency. 
In other instruments the grease is squeezed through a small 
opening by a definite force; here, again, results are most erratic 
and misleading. 

One reason probably the chief reason for these failures is 
that all greases have a peculiar "set," or "honeycomb" nature; 
once the grease has been handled, the honeycomb structure is 
broken up, and the grease becomes softer and more oily in appear- 
ance. To show this effect, the author forced a certain amount of 
a medium cup grease through a %-in. nozzle by the force of a 
28-lb. weight. The grease came through the first time in 126 
sec. On putting the same grease back in the test cup and repeat- 
ing the performance, the efflux times for the succeeding three tests 
were 47, 13, and 6 sec., respectively. 

Many engineers will have noticed that when working grease 
by the fingers and hands it becomes softer and softer. The author 
also tried various petroleum- jelly greases by a grease viscometer 
and found the results fairly consistent, presumably because these 
greases do not possess that peculiar structure characteristic of 
cup and other soap-containing greases. 

Capillarity. All lubricating oils have the property of rising 
into siphons or wicks made of wool or cotton, but the capillary 
power differs considerably for different oils. 

Railway and steamship companies, many of which employ to a 
large extent siphon lubrication or pad lubrication, find it very 
useful to compare lubricating oils for capillary power, as it is 


upon this property that their siphoning ability largely depends. 
Obviously, the best method is to test the oils in an actual box of 
the exact type used on the railway or steamer and to test the 
oils over the whole range of temperatures to which they may be 
exposed during service. 

The quality of the wool is important. Berlin wool, which is of 
a soft, loose texture, has greater siphoning ability than closely 
twisted worsted yarn. 

The siphoning ability of lubricating oil is influenced largely 
by the viscosity of the oil and by its nature, whether pure mineral 
or containing a percentage of fixed oil. The lower the viscosity 
the quicker will the oil siphon from the lubricator cup into the 

Emulsification. Circulation oils which are used in connection 
with steam turbines, enclosed-type steam engines, etc., come into 
contact with water and must not form an emulsion with it. 
Animal and vegetable oils emulsify quickly when churned 
together with water, so that it is out of the question to use these 
oils in circulation systems where there is danger of water's being 
present. Mineral lubricating oils have a low affinity for water, 
but experience has proved that this affinity is sufficiently strong 
in most of them to cause frequent trouble. The tendency to 
emulsify differs considerably for different oils, and it therefore 
becomes necessary to subject circulation oils to an emulsification 

This test may be carried out by shaking definite quantities 
of oil and water either by a reciprocating motion in a bottle or by 
churning the oil and water together by a paddle wheel revolving 
at high speed. The water may be distilled water, salt water, 
or a caustic-soda solution, according to the requirements that the 
oil has to meet. Marine-turbine oil, for example, must separate 
from salt water in such cases where a leakage of salt water into 
the system cannot easily be prevented. 

Where boilers prime, and boiler impurities are likely to be car- 
ried over into steam turbines or enclosed-type steam engines, 
it may be of interest to use a caustic-soda solution or even the 
boiler water itself when making the emulsification test. 

The test should be carried out at about 130F., which is the 
average temperature of circulation oils when in service, and the 
mixture should be allowed to settle at a similar temperature. 


Ferric oxide or iron salts have a most powerful emulsifying 
effect on circulation oils in the presence of water. If only a 
fraction of 1 per cent iron salts is added to the water used for 
the emulsification test nearly all oils will show a very consider- 
able percentage of sludge. It will appear that it is the presence 
of a quite small percentage of certain unstable hydrocarbons, sul- 
phur compounds, naphthene salts, etc., that causes emulsification. 

Filtered oils that have not been acid treated show less tendency 
to emulsification than acid-treated oils and should therefore be 
used in preference to the latter in the manufacture of circulation 
oils. When manufacturing heavy-viscosity circulation oils, it is 
often necessary to use an admixture of filtered cylinder oil or 
bright stock, which has not been treated with acid but merely 
filtered to remove unstable hydrocarbons, etc., and are therefore 
eminently suitable for the purpose. Well-filtered cylinder stocks 
have only a slight tendency to emulsification. 

Speaking generally, low-viscosity, low-specific-gravity oils give 
better service as circulation oils than heavy-viscosity, heavy- 
specific-gravity oils, because they separate more quickly from 
water, dirt, and other impurities. 

Attempts have been made to express the tendency of an oil to 
emulsify in terms of its " emulsification value, " an emulsification 
value of 98 per cent meaning that 98 per cent of oil separated 
out in the emulsification test, 2 per cent being retained in the 

Even if an oil shows great resistance to emulsification, it is 
desirable to know how rapidly the separation takes place. When 
in an emulsification test the mixture of oil and water is allowed 
to separate, some oils will separate out in a few minutes, whereas 
others may take half an hour or more, and yet the final separation 
may not show any formation of sludge. Obviously, quick 
separation is exceedingly important, as in most circulation oiling 
systems the oil is not given much time to free itself from 

An apparatus to determine the demulsibility, i.e., resistance 
to emulsification, of an oil and to express this quality by a figure 
has been devised by W. H. Herschel (described in U. S. Bureau of 
Standards, Bulletin 86). In this apparatus oil and water are 
churned by a paddle for 5 min., and a record taken of the time in 
minutes taken for separation. 


The demulsibility figure D is calculated as the rate of oil 
settling out per hour and is therefore expressed as 

D = 60 X -. 


where v = the volume of oil in cubic centimeters that has 

separated out. 

t = the time in minutes taken for the oil to separate out. 

The maximum demulsibility figure is 1,200; i.e., the entire 

volume of oil (20 cc.) separates out in 60 sec. If with a poor 

oil only 10 cc. separate out in, say, 15 min. ; the demulsibility 

value is 

60 X i5_40 

Surface Tension. There can be no doubt that the surface 
tension of an oil has some influence on the condition and strength 
of thin oil films in contact with metallic surfaces. Lubricating 
oils wet metallic surfaces, as their surface tensions are lower than 
those of metals. Differences in surface tension as between vari- 
ous lubricants will therefore mean different behavior as to their 
tendency to wet metallic surfaces, but the exact nature and 
importance of surface tension in connection with lubrication is 
still a practically unexplored subject. 


Acidity. Free acid in lubricating oils may be present as free 
mineral acid, petroleum acid, fatty acid, or rosin acid. 

a. Free sulphuric acid or other mineral acid which has been 
used in the refining of the oil. It is very rare nowadays to find 
any objectionable percentage of free acid from this source, but 
in the case of transformer and switch oils, it is of great importance 
that the percentage of mineral acid be exceptionally low, so that, 
whereas for ordinary purposes a percentage of 0.03 in terms 
of SO 3 may be permitted, the percentage in the case of trans- 
former oils must not exceed 0.01. 

6. Petroleum acid may be present in the original crude or 
may be produced during distillation and refining. Petroleum 
acids develop in circulation oils during continuous use, owing to 
oxidation. They are very weak in their action and affect no 


metals except lead and zinc; mineral oils usually contain less 
than 0.01 per cent of petroleum acids, but the presence of a larger 
percentage is not harmful as long as the percentage does not 
exceed 0.3 per cent in terms of SO 3 (used circulation oils). 

c. Free fatty acid is present only in lubricating oils that con- 
tain fixed oils. The percentage of free fatty acid in a fixed oil 
is not objectionable as long as it does not exceed 0.5 per cent in 
terms of SO 3. A higher percentage of acid is permissible in 
certain cutting oils. A mixture of fixed oil and mineral oil will, 
of course, contain proportionally less of free fatty acid the greater 
the percentage of mineral oil. 

When the content of free fatty acid is high in a lubricating oil, 
it has the effect of attacking the metallic surfaces with which the 
oil comes into contact. Metallic soaps are formed, which choke 
up the oil pipes and lubricating channels in the machinery. In 
contact with brass parts verdigris is formed. The softer metals 
like lead and zinc are very quickly attacked, and the effect is 
marked in bearings lined with white metal containing a high 
percentage of these metals. 

When oils containing fatty oils are stored in storage tanks or 
cabinets that are either unlined or merely galvanized, the free 
fatty acid attacks the metal surface, forming metallic soaps. It 
has been found, however, that tin is not attacked to any degree 
by the free fatty acid, and for this reason all oil cabinets and oil 
tanks should be tinned on surfaces in contact with the oil. This 
also applies to other parts of the cabinets, such as oil pumps and 

During continuous use, all oils containing fixed oils oxidize (air) 
and hydrolize (moisture), the result being the formation of free 
fatty acid and of sticky, gummy, varnishlike deposits, which 
may cause trouble. 

d. Rosin acids indicate the presence of rosin or rosin oil which 
is always objectionable in lubricating oils. 

Saponification Value. The saponification value is the number 
of grams of potash (KOH) required to saponify the fatty (vege- 
table or animal) constituents present in 1,000 g. of |he oil. The 
saponification value is therefore useful in determining the charac- 
ter and percentage of a fixed oil present in a mixture of fixed oil 
and mineral oil. When two or more grades of fixed oil are 
present it is difficult to identify them with certainty. 


Iodine Value. The iodine value is the number of grams of 
iodine absorbed by the unsaturated constituents present in 100 
g. of the oil. 

It has been mentioned that fixed oils have a great affinity for 
oxygen and that during continual use they will oxidize and form 
deposits. The iodine value is an indication of this tendency and 
is based on the fact that iodine will quickly combine with those 
ingredients in the oil which have a tendency to oxidize. 

As might be expected, the iodine value of drying oils like lin- 
seed oil is very high, whereas the iodine value of mineral lubri- 
cating oil is very low. Below are given typical iodine values for 
various oils: 

Drying oils, such as linseed oils Above 170 

Semidrying oils, such as cottonseed, ravison rape, fish, 

and whale oils From 100 to 170 

Nondrying oils, such as animal oils (except whale oil) and 

vegetable oils (except cottonseed and ravison rape) ... 50 to 100 

Scotch shale oil 0.890 23 

Russian mineral lubricating oils 7 

American mineral lubricating oils (paraffin base) 10 to 16 

The cause of the high iodine value of Scotch shale oil is the 
large percentage of unsaturated hydrocarbons present in it. 

Oxidation and Gumming. In order to get an idea of the tend- 
ency of lubricating oils to oxidize and gum, many tests have been 
devised; in one test, 1 g. of the oil is heated on a watch glass 
for a certain length of time at certain temperatures, after which 
the oil is examined. Another test measures the increase in weight, 
i.e., the amount of oxygen absorbed and the percentage of free 
fatty acid formed. All such tests are of value in comparing one 
oil with another. The iodine value, however, appears to be the 
nearest approach to a correct indication of the tendency of an 
oil to oxidize. 

Under " Color " it was mentioned that the color of an oil is 
due to the presence of unsaturated hydrocarbons. It is there- 
fore to be expected that oils dark in color are more easily oxidized 
than pale oils, and experience has proved this to be the case. 
Where machinery is exposed to sunlight, e.g., steam rollers and 
steam tractors, it has been found that red oils produce a tenacious 
dark-brown skin on the metal parts, whereas pale asphaltic or 
naphthenic-base oils have very much less tendency to form such 


Frequent complaints have also been made that machine parts 
in engine rooms become tarnished, also the bright parts of spindle 
frames in textile mills, unless very pale oils are used. This tar- 
nishing effect is very unsightly in the case of high-class machine 
tools. In order to avoid machine parts' becoming tarnished, it is 
therefore best to use pale-colored oils, either straight mineral or 
mixed with a small percentage of animal oil. The presence of the 
animal oil has a peculiar effect, making it quite easy to wipe the 
bright parts clean. Possibly, the free fatty acid present is 
helpful, preventing the film from forming, owing to a very 
slight corrosive action between the acid and the metal. An 
admixture of vegetable oil would increase the oxidizing tendency 
of the oil. 

For oils that are used in connection with transformers and 
air compressors, it is obvious that they must have the smallest 
possible tendency to combine with the air. Circulation oils (for 
steam turbines, etc.) are also in more or less contact with air and 
therefore subject to oxidation. It would therefore seem desirable 
to examine oils to be used for such purposes by subjecting them 
to art oxidation test on lines similar to the sludging test proposed 
by Michie for testing the sludging tendency of transformer oils. 
This test is carried out as follows: 

"One hundred cubic centimeters of the oil is placed in a 200-cc. 
flask and maintained at 150C. for 45 hr., during which period 
dry air is passed slowly through the oil at the rate of 0.066 cu.ft. 
per hour, a piece of copper with a total surface area 4}^ sq. in. 
being placed in the oil. 

"The amount of sludge found is then determined. " 

Ash. Ash is present in appreciable quantities only in lubri- 
cating oils that have been soap thickened or badly refined. 
Distilled mineral lubricating oils should not contain more than 
0.02 per cent of ash; and for undistilled oils like steam-cylinder 
oils, the ash should be less than 0.1 per cent. The ash may 
consist of iron rust from the still, or it may be alkali from the' 

Carbon Residue. Oils that during use are vaporized or 
burnt, as is the case with all oils used for internal-combustion 
engines, produce more or less carbon deposit. It is difficult to 
duplicate these conditions in a laboratory test, but it would seem 
desirable to have some kind of test for the tendency to carbonize, 


the results to be compared with actual practice in order to 
ascertain their value. 

One apparatus has been suggested by P. H. Conradson. This 
method is a modification of his original method and apparatus for 
carbon test and ash residue in petroleum lubricating oils. (See 
Proceedings Eighth International Congress of Applied Chemistry, 
New York, September, 1912, Vol. 1, page 131; also reprint in the 
Journal of Industrial and Engineering Chemistry, Vol. 4, No. 11, 
November, 1912.) 

Asphalt and Tar. It is seldom necessary to test lubricating 
oils for the presence of asphalt and tar, except in the case of dark 
cylinder oils, particularly those used for superheated steam. 

A distinction is made between hard asphalt and soft asphalt, 
the hard asphalt being the more objectionable of the two, as 
it will form hard, brittle carbonization deposits inside the 

Filtered cylinder oils contain less asphalt than dark cylinder 
oils and are therefore to be preferred in all such cases where 
carbonization ordinarily may be expected to take place. 

Oiliness. The property in a lubricant that causes it to adhere 
to metallic surfaces is generally referred to as its oiliness. 

No means have as yet been devised by which the power of a 
lubricant to adhere to metallic surfaces can be directly measured ; 
if lubricated surfaces are pulled apart, the lubricating film itself 
is severed, but the lubricant still adheres to both surfaces, so that 
it is only the cohesion of the film that can be determined in this 
manner. This property is of no value under fluid friction condi- 
tions, when the viscosity of the oil is the only factor of impor- 
tance, but is the most important property under boundary 
lubrication conditions, when the film of lubricant is so thin that 
no part of it lies outside the range of the molecular forces which 
attract the bearing surfaces mutually. These molecular forces 
make themselves felt at a considerable distance from the surface 
molecules (according to Hardy, up to a distance of 0.0016 in.). 
The oil molecules are therefore adsorbed with a considerable 
force by the metal molecules of the sliding surface ; they penetrate 
the surface to a certain depth; and, according to the greater or 
lesser chemical activity (content of surface-active polar bodies), 
they combine more or less intimately and/or cling more or less 
tenaciously to the metal surface molecules. 


Under these conditions, the viscosity of the oil is immaterial; 
it is the chemical nature of the oil its oiliness that matters. 
Two oils may have the same viscosity, and the oiliness be much 
greater in the one than in the other. It is common knowledge 
that all fixed oils, or mixtures of mineral oils and fixed oils, 
possess greater oiliness than straight mineral oils. From the 
writer's experience it seems also certain that a mixture of low- 
viscosity distilled mineral lubricating oil and filtered cylinder 
stock has greater oiliness than a distilled mineral lubricating 
oil of the same viscosity as the mixture. 

That distilled lubricating oils are improved in oiliness by the 
admixture of fixed oil or filtered cylinder stock may be explained 
by the fact that molecules of the fatty oil or cylinder stock 
adhere to the metallic surfaces in preference to the molecules of 
the distilled mineral oils ; such coating of the surfaces with strongly 
adhering molecules explains why it is possible with blended oils 
to sustain almost as great pressures as if the fixed oil or filtered 
cylinder stock were used alone. 

It is a remarkable fact that great oiliness in a lubricant is 
produced by the presence of a quite small amount of a "very 
oily" lubricant. For example, compounding a spindle oil with 
6 per cent of a fatty oil produces an oil that for many purposes is 
just as good as if 10 per cent of fatty oil were used. 

As long as there are sufficient very oily molecules present to 
coat the frictional surface, the oiliness cannot be much improved 
by further addition of the very oily lubricant. 

This is brought out very clearly by experiments by J. H. Hyde 
(Report of Lubrication and Lubricants Inquiry Committee, 1920, 
pages 63 and 64) and even more so by his experiments reported 
in Engineering for June 10, 1921, pages 708-709, where it is 
also proved that small additions of fatty acid have a greater 
influence than neutral fatty oils in improving the oiliness of 
mineral oil. 

Interesting results showing the value of fatty acids as lubri- 
cants were given in a paper read on Feb. 2, 1920, by Henry M. 
Wells and James E. Southcombe before the Society of Chemical 
Industry, London. 

The author has used the addition of from 2 to 10 per cent of 
fatty acid in various internal-combustion engine oils and others 
as far back as 1908 and found the acids more active than neutral 


fatty oils; only they must be used with a great deal of judgment 
to avoid trouble. 

Experience shows that the oiliness is greatest for fatty acids, 
smaller for fatty oils, and smallest for mineral oils, although 
varying within fairly large limits inside each group. 

It is now generally recognized that the oiliness of a lubricant 
is determined only. to an insignificant degree by its physical 
properties but chiefly by its chemical properties or, rather, the 
chemical properties of its very oily constituents which need 
amount to only a small fraction of the bulk of the lubricant. 

Extreme-pressure lubricants have been developed during the 
last few years and are finding an increasing use for lubrication of 
highly loaded tooth gears, for heavily loaded bearings, for wire- 
drawing, deep pressing of metal, etc. 

A method of testing extreme-pressure lubricants has been 
developed by the Engine Testing Station at Delft, Netherlands, 
belonging to the Royal Dutch Shell Company. It consists 
essentially of four J^-in. balls in the form of a pyramid; the three 
lower balls are placed in a cup, into which the oil to be tested is 
poured. The top ball is rotated, and the cup with the three balls 
is forced against the top one. The cup will now endeavor to 
revolve, and the torque is measured. 

Two standard tests are made: (1) a 1-min. test under heavy 
load; (2) an 8-hr, test using lower loads. 

The torque during the tests and the wear of the balls are 
measured. The wear diagram obtained from the short-duration 
test is said to be of great importance for evaluating the load- 
carrying capacity of the lubricants. 

W. J. D. van Dyck 1 after describing this test method, said that 
in this apparatus a good mineral oil cannot stand a much higher 
load than 200 lb., and a fatty oil not much more than 250 Ib. 
An extreme-pressure lubricant on a sulphur base showed far 
better results, while a lubricant on a combined chlorine and sul- 
phur base would withstand a load ten times that of mineral oil. 
Van Dyck emphasized that the properties of load capacity and 
wear measured in this manner are not conclusive for evaluating 
extreme-pressure lubricants in practice. Other factors, such as 
stability in storage and in use; the absence of corrosion on the 
different materials with which the oil might come into contact, 

1 Engineer, Oct. 11, 1935, p. 382. 


both when the lubricant is fresh and when it is used for long 
periods; and stability against water, must all be taken into 
account, as in the case of normal lubricants. 

Impurities. The impurities most frequently met with in 
lubricating oils are dirt, glue, and water. 

Dirt. Dirt is easily detected when the oil is transparent. 
It is more difficult in the case of dark oils, such as cylinder oils. 
The best way of testing a lubricating oil for dirt is to draw a 
few gallons of oil from the bottom of the barrel or the tank in 
which the oil is stored and then strain the oil through muslin or 
silk cloth. Anything that remains on the cloth can be freed from 
oil by treatment with gasoline, and it is then generally easy 
enough by the aid of a magnifying glass or perhaps even with the 
naked eye to judge what the impurities are. 

If metallic iron in the form of iron scale is present (from the 
steel drum or barrel), a magnet will detect it, the small particles 
being drawn up at the approach of the magnet; or the metallic 
ingredients can be identified chemically. 

Cotton waste, small pieces of wood, etc., are easily recogniz- 
able, and it is not unusual for them to find their way into the oil 
when the barrel is being broached. 

The bung should be loosened by striking the staves with a 
mallet, and it should never be removed by the use of an augur. 

Glue. Glue is used for impregnating the inside of wooden 
barrels and serves two purposes: (1) preventing oil leakage 
through the wooden staves; (2) preventing moisture from entering 
the oil. 

The importance of the first-mentioned object is obvious, and 
the desirability of preventing moisture from entering the oil is 
explained in the next paragraph under " water. 7 ' 

Sometimes large quantities of glue are found in a barrel because 
the barrel has not been properly drained of the hot liquid glue 
during the gluing process. The glue will, however, not mix with 
the oil except in the presence of moisture and can always be 
detected easily by the consumer, when the oil is being strained 
before use. If the glue is not detected, the results may be 
disastrous, as it will cause excessive heating and wear and 
develop sticky deposits in lubricators, in circulation oiling sys- 
tems, and in oil pipes; it will cause irregular working of lubricators, 
especially those having fine openings, as hydrostatic-displacement 


cylinder lubricators. When such fine openings become choked, 
steam valves and pistons cry out for oil if the trouble is not 
observed in time. 

Water. Water gives an oil a cloudy appearance, and its 
presence is therefore easily perceived in oils, which in dry condi- 
tion are transparent. 

When the oil is heated to a few degrees above 212F. it will 
soon become transparent, if the cloudiness is due to water; and 
if more than a trace of moisture is present, it will partly evaporate 
and partly separate out as visible drops of water at the bottom. 

Mineral oils are more easily clarified than oils containing 
fixed oils, as the latter have a strong affinity for water and easily 
become emulsified. 

The presence of even a trace of moisture is very detrimental 
in transformer and switch oils. A simple test (apart from testing 
dielectric strength or specific resistance) is the hot-iron test. 
An 8-oz. bottle is half filled with transformer oil; an iron rod, 
say J4 in. in diameter, is heated for about ^ in. to a dull red heat 
and slowly lowered into the oil. If more than 0.01 per cent of 
water is present, the tiny particles of water will suddenly turn 
into steam with a crackling noise ; if no water is present, there will 
be only a slight hissing noise from the oil vapors. 

The presence of a slight amount of moisture in oils, other than 
transformer and switch oils, is not detrimental, so far as the 
influence of the water itself is concerned; and yet, in nearly every 
case where the oil is moist, more or less trouble is experienced. 
Ring spindles and other textile spindles rust and run warm; 
internal-combustion engines develop an excessive amount of 
carbon deposit; the pistons heat up and wear rapidly; the oil is 
reported to be " thinner than usual" (because the excessive heat- 
ing of the oil film thins the oil); and the oil comes out of the 
pistons and bearings in a chocolate-colored or blackened, dirty 

This very remarkable effect of the presence of small amounts 
of moisture is explained by the fact that moisture nearly always 
gets into the oil through exposure of the wooden barrels to the 
weather. During warm and rainy weather the staves expand 
and absorb moisture; during nights they contract, and the effect 
of such alternate expansion and contraction is that moisture 
gets through to the inside of the staves, loosens and dissolves 


some of the glue coating, and spreads it throughout the contents 
of the barrel. This is the most dangerous form in which glue 
can be present in the oil, and it is usually the glue that causes 
lubrication troubles, more so than the water. Wooden barrels, 
when in transit, should therefore preferably be covered with 
tarpaulins and should be stored under cover in a dry place. 
When barrels are stored out-of-doors from lack of space under 
roof, they should be covered with waterproof covering or stored 
on their sides; when stored on end, the moisture collects over the 
staves, and there is a greater likelihood of the water's getting 
inside than when they are stored on their sides. 


Mechanical Testing Machines. As the usual physical and 
chemical tests of lubricants do not always definitely indicate 
whether one oil will be more satisfactory than another for certain 
machines, many investigators have designed friction-testing 
machines with a view to comparing the lubricating properties 
of different oils. There is a great variety of these machines, 
chiefly for testing bearing oils, and they have been extremely 
useful in discovering important laws of friction and in comparing 
the efficiency of different lubricating systems. The results of 
such experimental work have been of interest to oil manufacturers 
and lubrication engineers, but from an oil consumer's point 
of view they are, speaking generally, of no value so far as the 
selection of suitable oils is concerned. 

The difficulty is that the testing machine has only one bearing, 
usually with beautifully finished rubbing surfaces and operated 
under conditions of oil feed, pressure, speed, and temperature 
quite different from practical conditions. In most works there 
are such a variety of bearings that it is quite impossible to repro- 
duce all these conditions on the one bearing of a testing machine. 

The following two examples may prove instructive : 

Example 1. A certain government had a Lahmeyer oil- 
testing' machine, with which all of the oils offered by various firms 
were tested. The oils were intended to be used on the propelling 
machinery of warships. 

In the Lahmeyer testing machine two heavy flywheels are 
carried, one on each end of a shaft; the shaft is supported by a 


central ring oiling bearing, which serves for testing the oil. The 
machine is driven by an electric motor, which can be connected 
to the flywheel shaft by a pin coupling. The method of testing 
is as follows: 

The bearing is supplied with the oil to be tested. The motor is 
started, and the flywheel rotated at full speed 1,500 to 1,700 
r.p.m. The motor is then uncoupled, and the time noted that 
elapses before the flywheel comes to rest. 

The longer the time taken by the shaft to come to rest the better 
is the quality of the oil supposed to be, and this is true for this 
particular bearing. 

Before the next sample of oil is tested the bearing is quicl$|y 
cleaned by benzine passed through it, and it is dried out with an 
air current. 

It will be understood that an oil manufactured to meet the 
conditions of a high-speed ring-lubricated bearing will give the 
best results when tested on this machine. 

In order to convince the government in question as to the 
futility of testing oils in this manner, a good dynamo oil was 
submitted, and it was found that the shaft revolved three times 
as long as with the marine oil, which in actual practice gave the 
best results. It was obvious to everyone concerned that the 
dynamo oil was absolutely unsuitable for the work required. 

Example 2. One of the best oil-testing machines on the market 
is Thurston's machine. The machine consists of a shaft sup- 
ported by two bearings; the shaft at one end has an overhanging 
bearing fitted with two brasses, on which the oil is tested. Sus- 
pended from the bearing is a hollow pendulum containing a 
spring, by means of which a certain bearing pressure may be 
maintained. When the shaft revolves, the oil film interposed 
between the shaft and the brasses causes the pendulum to swing 
outward, and it remains in a certain position according to the 
oil in use. 

The less the outswing the less is the coefficient of friction 
and the better the oil, for this particular bearing and for the 
particular conditions prevailing. 

When tests were carried out to find out which was the most 
suitable oil for shafting bearings running at a certain speed and 
bearing pressure, a Thurston testing machine was made to run 
under as nearly as possible similar conditions; it was found, 


however, when testing different oils, that the coefficient of friction 
was the least for pure kerosene, which would, of course, be useless 
for the lubrication of shafting bearings. 

This result will be easily understood when one takes into con- 
sideration the fact that shafting in actual practice is always more 
or less out of line and that the bearing surfaces are never perfectly 
smooth. The pressure, therefore, will not distribute itself so 
uniformly over the entire bearing surfaces, as will be the case 
with the bearing of Thurston's oil tester. 

The limitations of testing machines are now beginning to be 
generally recognized; it is only where, as in the case of railways, 
a great many bearings are alike and operating under similar 
conditions that it seems at all worth while to attempt the con- 
struction of a testing machine; even then there is ample evidence 
that variations in the results obtained with the same oil in use 
may easily amount to 50 and rarely fall below 10 per cent. 

The author feels that, from the consumer's point of view, the 
coefficient of friction of various oils as determined by a testing 
machine is not of much use; the oil that gives the least friction 
on the testing machine may often prove to be unsuitable in 
actual use. 

In the foregoing no reference has been made to testing machines 
for testing oils for internal lubrication of steam cylinders, gas 
engines, etc. Not a few attempts have been made in these 
directions, but, as far as the author knows, the results have been 
of doubtful value, if not altogether misleading. It must be 
kept in mind that in the internal lubrication of, for example, 
steam cylinders and internal-combustion engine cylinders, the 
lubrication is nearly always imperfect and subject to so many 
influencing factors that it is much more difficult to reproduce 
the conditions in a testing machine than in the case of bearings. 
Besides, the value of a lubricant often becomes apparent only 
after several weeks or months of use ; such properties as tendency 
to carbonize, emulsify, oxidize, etc., may become of paramount 
importance, as compared with the friction-reducing properties 
of the oil, as will be made clear later on in the various chapters 
devoted to the different kinds of engines. 

Works Tests. The author has come to the conclusion, and 
most lubrication engineers will, he feels certain) agree with him, 
that the only reliable way of testing lubricants is to test them under 



actual working conditions, by applying them to the machinery 
upon which they are to be used, and watch the results. 

FIG. 2. Taking the temperature of a ring-oiling bearing. 

Temperature Tests. The simplest method of comparing two 
oils is to compare the frictional temperature rise of typical 

FIG. 3. Taking the temperature of a pedestal bearing. 

bearings, using first one oil and then the other. Any difference 
in quality or suitability between the two oils will be shown by 



a different frictional rise in temperature above the surrounding 
air temperature. The difference in temperature between a bear- 
ing and the air close to it will remain the same, independent of 
the air temperature, as long as the same oil is in use. 

Figures 2 and 3 show the method of taking the oil temperature 
of a ring-oiling bearing and a pedestal bearing. In the first case 
the thermometer bulb is immersed in the oil; in the second the 
bulb is covered with a lump of fairly stiff grease or putty, so that 
the bulb may be held in contact with the metal and as accurately 
as possible record the correct temperature. 

Below is given a typical example of a temperature test on a 
ring-oiling bearing, a viscous oil being compared with an oil of 
the correct light body: 

Temperature of 

Temperature of 

Frictional rise 


engine room, 

dynamo hearing, 

in temperature, 

JL lalllj 







9.45 A.M. 




10.0 A.M. 




10.30 A.M. 




11.0 A.M. 

Change made to low-visconity oil 

1.0 P.M. 




1.30 P.M. 




2.0 P.M. 




3.0 P.M. 




4.0 P.M. 




It sometimes takes several weeks before the minimum tem- 
perature is reached, especially when there is a great difference 
between the two oils. 

Special thermometers are used for taking spindle-rail tempera- 
tures; one method is to fix a shallow box to the rail; the bottom 
of the box near the rail has a long slit into which the thermometer 
is fixed, the bulb being pressed lightly against the rail; the box 
has a hinged lid, which is lifted only long enough for the tempera- 
ture to be read. 

Temperature tests are extremely useful for comparing oils 
in actual use, and the tests should be repeated from time to time 
with a view to checking the quality of the oils in use. If the 
mechanical conditions do not change, the rise in temperature 



of the bearings above the surrounding air should remain very 
nearly constant. 

In order that reliable temperature readings may be taken, 
quick-registering and accurate thermometers should be used. 
Most engineers' thermometers are sluggish and liable to be frac- 
tured when carried in the pocket or dropped to the ground. The 

FIG. 4. Engineer's 

FIG. 5. Thomson's engineer's 

author's staff of engineers broke so many thermometers of the 
type illustrated in Fig. 4 that he designed a special thermometer 
and case, as shown in Fig. 5. The thermometer head is flexibly 
secured in the cap, which fits into the case with a bayonet joint, 
and when in position the bulb of the thermometer is kept central 
out of contact with the case by means of a spring pad. When 
the thermometer is carried in the pocket, it cannot be broken, 
and it is prevented from dropping out by the safety pin fastened 



to the clothing. The introduction of this thermometer reduced 
the number of breakages practically to nil. 

Dynamometer Tests. Several dynamometers, such as Emer- 
son's and Bailey's, are employed for measuring the power con- 
sumed by individual machines, such as spinning frames, but only 
for small horsepowers. 

FIG. 6. Emerson's dynamometer. 

Emerson's machine (Fig. 6) is the first instrument that was 
ever used in the oil business for the purpose of showing the value 
of good lubrication. It is fixed to the driving shaft outside the 
loose pulley. The pull of the belt goes through the arms, which 
pull back levers, just like an ordinary weighing scale, and the 
pointer shows the number of pounds exerted. The diameter of 
the wheel is 2 ft. The speed is measured in r.p.m., and the horse- 
power is calculated from the formula 

net weight X 2 X r.p.m. 

Horsepower = 



These instruments are so finely adjusted that if two or three 
spindles are stopped by hand, the pointer immediately registers 
the increased friction. 

Electrical Tests. Where a machine or a group of machines is 
driven by electric motors, it is a simple matter to record the 
power consumption. But apart from the electrical measure- 
ments (volts, amperes, kilowatts or B.t.u. per hour, as the case 
may be) it is desirable or necessary to record, as with the spinning- 
frame tests, the temperature and relative humidity of the air; 
the speeds of motor, shafting, and machines; and the frictional 
temperatures of important bearings, all with a view to getting 
as complete indications as possible of the alterations caused by a 
change in lubricants or lubricating methods. 

Steam-engine Tests. To record the power consumption of a 
factory or mill by means of indicator diagrams taken from the 
engine requires extreme care for the purpose of making a compari- 
son between two sets of lubricating conditions. The load always 
varies, even under the most ideal conditions; the governor is 
continuously altering the amount of steam admitted, and dia- 
grams taken quickly after one another may differ appreciably. 

The only accurate method is to take a great number of indica- 
tor diagrams (preferably on Tuesdays, Wednesdays, or Thurs- 
days) at regular working intervals, say every 10 or 15 min. during, 
say, 4 working hours; the indicator-pencil motions may be fitted 
with magnets, so that by the closing of a switch all diagrams 
can be taken simultaneously. 

An accurate note must be made of machines stopped in the 
mill; if, for example, a machine consuming 8 hp. is stopped for 
half an hour, the equivalent value over the 4 hr. is 1 hp. If 
the values of all such stoppages are added together and amount 
to 17 hp., and the average indicated horsepower, calculated 
from all diagrams, is 805 hp., then it may be assumed that the 
mill would consume an average of 822 hp. if all the machines 
had been working continuously. 

If on the comparative test, say 3 months later, the average 
power with other oils works out at 742 hp. and the value for 
machines stopped is 14 hp., then the comparative power value 
with the new oils is 754 hp., which represents a saving of 68 
hp., assuming that the conditions as regards temperature, rela- 
tive humidity, etc., are similar. 


Gas-engine Tests. The usual particulars should be recorded; 
the gas consumption can be taken when a gas meter is installed 
and should be reduced to a basis of 32F. gas temperature and 
28 in. mercury barometric pressure, so that the amounts of gas 
consumed on both tests may be made comparable. The tem- 
perature and pressure of the gas should therefore be recorded, 
also the calorific value of the gas. The temperature of water 
inlet and outlet for the water jacket, temperature of intake air, 
the position of air intake on engine (if variable), and the number 
of actual explosions per minute should be recorded, as they may 
prove important in comparing the results of two sets of oils. 

Where the gas consumption cannot be recorded, indicator 
diagrams may be taken, from which the power consumption can 
be calculated. 

"Free-revolution" Tests. An approximate comparison be- 
tween two sets of lubrication conditions may be made by running 
a number of transmission shafting, countershafting, and machines 
idle at normal speed and then suddenly shutting off steam, gas, 
electricity, or whatever moving power is employed. The prime 
mover (steam engine, gas engine, electric motor, etc.) will con- 
tinue to operate for a certain number of revolutions and for a 
certain length of time. By improving the lubrication, the prime 
mover will run for a longer period and a greater number of "free 
revolutions " before it comes to a standstill. This method is not 
very scientific but is simple to carry out and often useful. 

Similar effects are noticed on spinning frames; with improved 
lubrication they run for a longer time when the belts are thrown 
on to the loose pulleys. In the same way, the driver of a hoisting 
engine finds that he opens his throttle later and closes it earlier 
when the lubrication of valves and cylinders is improved. It 
will generally be found that engine attendants or machine opera- 
tors who have handled their machines for a long time have some 
way of judging the state of lubrication efficiency. They know 
at once if there is a change, although many of them do not 
know how to express themselves in technical terms. 

General Remarks. As to selecting a suitable part of a factory 
for a test, it is difficult to give general rules. It will often be 
found that the engineer of an up-to-date works has a favorite 
piece of plant on which he makes all his trials and tests. Such 
a place should always be given preference, provided, of course, 


that it meets all requirements, as the engineer will be more famil- 
iar with the running of such machinery and will the more readily 
notice any improvement achieved by changing the lubricant. 

Where it is possible, a compact group of machines should be 
chosen. It is desirable that the group be compact, so that the 
whole of the plant may be under observation of the operator 
while running; the stoppage of a machine, the breaking of a belt, 
or the heating of a bearing can be seen at once, a note made of 
the time when the machine is put out of action, and allowance 
made for it in the final results. 

It must not be forgotten that a considerable time must usually 
be allowed between two comparative tests, to ensure that condi- 
tions with the new lubricants in use have become uniform. 
Where speeds are high, and both sets of oils are pure mineral in 
character, a few weeks will be sufficient; but where speeds are 
lower and pressures heavier, and particularly if the oils in the 
first set are compounded and in the second set straight mineral, 
or if there is a great difference in viscosities, the author has found 
that the change in power consumption may easily take three 
months to be fully accomplished. 



Without friction, life in the various forms in which we are 
acquainted with it would exist only for a very short time. Any 
moving mass would retain and continue its motion. If it were 
sliding down an incline and accelerating, it would reach another 
incline and rise to a certain height, then move to another position 
at the same height above sea level, aftd continue without ever 
coming to rest. 

Everything except the solid rocky formations would start slid- 
ing. Towns and cities would be swept away with the country; 
steamers on the open sea, at the moment friction ceased to be, 
would not be able to accelerate or decrease their speed, as the fric- 
tion between the propeller and the water and between the particles 
of water themselves would be nonexistent. Sailing ships would 
be in the same plight, as the wind would have no effect on the 
sails. Locomotives would not be able to move, as there would be 
no rail friction. 

Friction may be defined as the resistance created by the surface 
of o'ne body moving over that of another. If no lubricant 
is introduced between the surfaces, the friction may be termed 
" solid." If there were nothing but solid friction, very little 
machinery could be kept in operation: fast-going steamers and 
railway expresses would be unknown ; and only the crudest forms 
of slow-running machinery could be operated. 

Solid Friction. All surfaces are more or less rough; even sur- 
faces that are well machined and polished show under the micro- 
scope small projections and depressions. It is the interlocking 
of these minute projections that cause solid friction when two 
unlubricated surfaces are pressed together and move relative to 
one another. ' 

When the rubbing surfaces are very smooth and in intimate 
contact, an additional resistance to motion is created by adhesion 
between the surfaces caused by molecular attraction. This 



adhesive force is shown by Johnson's Swedish limit gauges 
used in many engineering works. When two or more of these 
gauges are brought into close contact, they adhere with a force 
several times that of the atmospheric pressure, and it is difficult 
to slide one surface over another, notwithstanding the absence of 
external pressure. Speaking generally, the laws of solid friction 
are as follows : The f rictional resistance with solid friction is 

1. Directly proportional to the total pressure between the surfaces. 

2. Independent of the rubbing speed of the surfaces at low speeds but 
decreases at very high speeds. 

3. Independent of the areas of the surfaces. 

4. Dependent to a considerable extent on the roughness and hardness 
of the surfaces. 

These laws apply whether the motion is rolling or sliding; they 
apply, therefore, also to ball and roller bearings. 

That the friction decreases at high speeds is well illustrated by 
the greatly diminished brake effect of automobile brakes at very 
high speeds. The action of the brakes may become so reduced 
that it may not be possible to regain control of the car when going 
down a steep hill. 

Contaminated Surfaces. It is an important fact that surfaces 
are never perfectly clean. Chemically clean surfaces soon abrade 
and weld themselves together when rubbing over one another; 
fortunately, all surfaces are covered with what may be called 
contamination films of a more or less greasy nature; these films 
are due to the action of air, moisture, dust, and impurities on the 
surfaces, and they help to some extent in preventing abrasion 
at any rate under low-pressure conditions in fact, they act very 
much like thin lubricating films. Archbutt and Deeley mention 
the following experiment to illustrate the effect of contamination : 

A smooth file passed over a freshly prepared clean surface will be 
found to cut well even when only gently pressed against the metal; 
but if the hand be passed over the metallic surface, the film of grease 
therefore deposited will so lubricate it that considerably greater pressure 
on the file is now needed to cause it to cut. 

Owing to the surface irregularities of the rubbing surfaces, 
wear takes place, the softer surface being more rapidly abraded 
than the harder. The wear and friction are much less for hard 
and smooth than for soft and rough surfaces. 


Surfaces of exactly the same material are more inclined to seize 
and weld than dissimilar surfaces; this is the reason why materials 
of different hardness and composition are used for all rubbing 
surfaces, e.g., a steel journal in a white-metaled bearing and soft 
cast-iron piston rings against a harder cast-iron cylinder. 

Although the friction between solid surfaces is independent of 
the area in contact, the wear is obviously the greater the smaller 
the area, because of the greater pressure per square inch. 

By the introduction of a suitable third medium between the 
frictional surfaces, a medium that may be solid (such as graphite, 
talc, or white lead) or of an oily nature (such as lubricating grease 
or lubricating oils), the solid friction may be partially or wholly 
eliminated, and, with the latter mediums, replaced with "soft 
solid" or fluid friction. Roller bearings and ball bearings are 
excepted in this connection. 

Fluid Friction. The object of all lubrication is that the lubri- 
cant should attach itself to the rubbing surfaces and form a 
film between them, which, under the conditions of speed, pressure, 
and temperature prevailing, will not be squeezed out but will 
keep the frictional surfaces apart. This object is not often 
attained, except in high-speed bearings, e.g., stream-fed bearings 
lubricated by a circulation oiling system, as in steam turbines 
and high-speed steam engines; many ring-oiling bearings; Michell 
bearings; and Nomy bearings. 

In bearings thus perfectly lubricated the " rubbing" surfaces 
never touch one another, and the friction is entirely dependent on 
the lubricant. The laws governing fluid friction are totally 
different from the laws for solid friction and may be summarized 
as follows: The frictional resistance with fluid friction 

1. Is independent of the pressure between the surfaces. 

2. Increases with speed of rubbing surfaces. 

3. Increases with area of rubbing surfaces. 

4. Is independent of the condition of the rubbing surfaces, or the materials 
of which they are composed. 

5. Depends entirely on the viscosity of the lubricant at the working tem- 
perature of the oil film. 

If the frictional resistance is F, and the total pressur^ between 
the rubbing surfaces P, then the friction equals P multiplied by 
the coefficient of friction C, i.e.: 

F = C XP 



The coefficient of friction for unlubricated surfaces ranges 
from 0.1 to 0.4, but with fluid friction the coefficient of friction 
ranges from 0.002 to 0.01 according to the viscosity of the oil. 
It is therefore worth while, wherever possible, to design bearings 
so that fluid friction, or a condition approaching fluid friction, can 
be brought about. 

Boundary-lubricated Surfaces. Under conditions of low speed 
and high pressure it is impossible or extremely difficult to obtain 
perfect film formation, nor is it possible in the great majority of 
bearings, which are not stream fed but supplied with only a 
limited amount of oil per minute, to produce anything approach- 
ing perfect film formation. The surfaces accordingly are in an 
imperfectly lubricated or semilubricated condition -boundary 
lubrication for which the coefficient of friction will range from 
0.01 to 0.10 according to whether the surfaces are very poorly 
lubricated approaching the condition of unlubricated surfaces 
or fairly well lubricated approaching the condition of perfectly 
lubricated surfaces. 

There Are No Definite Laws Governing the Lubrication of 
Boundary-lubricated Surfaces. The frictional resistance is com- 
posed partly of solid friction and partly of fluid friction, and the 
more the solid friction predominates the more important is the 
property known as oiliness, and the less important the viscosity of 
the lubricant. The object of lubrication of such surfaces is to 
make the best possible compromise between reduction of wear 
and reduction of fluid friction. For conditions of low pressure 
and high speed, the reduction of fluid friction is usually the most 
important point to consider and demands low-viscosity oils of 
great oiliness; whereas for conditions of high pressure and low 
speed, the reduction of wear must be given prime consideration 
and therefore calls for viscous oils of great oiliness. 

In ball and roller bearings the friction is usually not influenced 
by lubrication and is lower than the friction in even the best 
lubricated plain bearings. 

On p. 85 are given approximate values for the coefficient of 
friction for the sake of comparison. 



Coefficient oi 


Condition of surfaces 



Unlubricated or very poorly lubricated surfaces . . 
Boundary-lubricated surfaces 

0.1 toO,4 
0.01 to 0.10 


Perfectly lubricated surfaces 

0.002 to 0.01 


Surfaces in rolling contact: 
Ball bearings 

0.001 to 003 


Roller bearings 

002 to 007 


Static Coefficient of Friction. The values given above for the 
coefficient of friction are the kinetic values, applying to surfaces 
in motion. When surfaces have been at rest for some time, the 
oil film is more or less completely squeezed out, and a certain 
amount of metallic contact takes place. As a result, the start- 
ing effort, when the surfaces are again brought into motion, is 
much greater than the running effort; in fact, the static coefficient 
of friction usually approximates the values for solid friction. 

When the speed of the rubbing surfaces is very low, the kinetic 
coefficient of friction may be even higher than the static value, 
as there is added to the solid friction the resistance caused by 
the presence of a lubricant, it being understood that the speed of 
rubbing is too low to allow the lubricant to produce any appre- 
ciable separation of the rubbing surfaces. As the speed increases 
and the lubricant begins to produce a film, the solid friction 
quickly decreases, and the kinetic coefficient of friction is likewise 
reduced, until perfect film formation is brought about. 

The high values for the static coefficient of friction explain 
the great effort often required to start engines or machinery from 
rest and form one of the chief reasons why ball and roller bearings 
are used, as with surfaces in rolling contact there is practically 
no difference between the static and the kinetic coefficient of 

The static coefficient of friction will obviously depend on 


1. The condition and hardness of the surfaces, being lower for hard and 
smooth surfaces than for soft and rough surfaces. 

2. The pressure between the surfaces; the greater the pressure the more 
effectively is the lubricant squeezed out. 


3. The length of time that the surfaces have been at rest; the longer the time 
the greater chance has the pressure of displacing the lubricant. 

4. The nature of the lubricant. 

Solid lubricants like graphite are not displaced, so that in 
bearings lubricated entirely by solid lubricants the static and 
kinetic coefficients of friction (within reasonable limits) are very 
similar. Semisolid lubricants cannot be entirely displaced by 
pressure during a period of rest ; this is an advantage as compared 
with oils which occasionally may be of importance. Mineral 
oils are almost completely displaced, but experience proves that 
fixed oils or mineral oils compounded with fixed oil or oils contain- 
ing colloidal graphite leave a better film in between the surfaces 
and that therefore the static coefficient of friction with the latter 
oils is considerably less than with straight mineral oils. As a 
result, not only is the starting effort reduced but also the wear 
caused by metallic abrasion during the initial moments of starting. 

Temperature and Character of Frictional Surfaces. Bowden 
and Ridler (Laboratory of Physical Chemistry, Cambridge) have 
shown that the temperature at the interface between sliding 
metals may reach a very high value, say exceeding 1000C., with 
moderate sliding speeds and loads in the case of polished metals. 

They repeated the experiments with various lubricants under 
boundary-lubrication conditions and again found high surface 
temperatures. For example, with a polished metal surface 
lubricated with Castrol XL and running smoothly with a low 
coefficient of friction, a surface temperature of over 600C. was 
recorded at the sliding surface and yet the mass of metal was 
at room temperature, and there was no evidence of heating. 

The results have an important bearing on the theory and prac- 
tice of lubrication. The high local temperature may cause 
decomposition and volatilization of the lubricant and is thus an 
important cause of the breakdown of lubricating oils. 

Bowden and Ridler's experiments throw some light upon what 
engineers call "skin." We talk about bearings or engine cylin- 
ders' developing a surface skin, a good working surface, a 
mirror-like surface, etc. ; we know that every change in the lubri- 
cating conditions (changing oil or reducing the feed, etc.) alters 
the skin; that if this change takes place too suddenly, the skin 
may break, and trouble result. 


It is evident that if the microscopic surface irregularities are 
subject to temperatures of the magnitude recently referred to, 
structural changes take place in the bearing surfaces until a 
surface skin has been developed; and as this obviously takes time, 
every change in oil or lubricating conditions must be made with 
reasonable care and precaution. 

When during operation a surface skin is formed, this skin is 
very valuable, as it will sustain a much higher load than the 
normal load under which it was formed. Once such a skin is 
broken, it is important that it, or the oil film, be of such a char- 
acter that it easily re-forms, and this will, of course, depend upon 
the oiliness of the lubricant (see page 66) and the character of the 
metal surfaces. 

Graphoided surfaces which are produced when using lubricants 
mixed with colloidal graphite also appear to possess this character 
because the graphite, when embedded in the pores of the metal, 
cannot be squeezed out or easily displaced; and as the colloidal 
graphite has a great affinity for oil, the oil film will rather readily 
re-form over a graphoided surface. 


The main types of lubricators and lubricating appliances are 
described under " Bearings." It will carry the author too far to 
elaborate further on the many types and constructions of lubri- 
cators in existence; he hopes that sufficient is said under " Bear- 
ings" to convey his views on the merits or demerits of the 
various principles involved. 

As, however, mechanically operated lubricators are coming 
much into prominence, and as the author has taken a particular 
interest in these appliances, he feels that a critical review of the 
main types may prove useful. 

Mechanically operated lubricators are now widely used for 
delivering a small or moderate supply of oil automatically and 
at a uniform rate of feeding, against a pressure ranging from a 
few to as much as 1,000 Ib. per square inch. 

Mechanical lubricators are used for feeding oil to the cylinders 
and valves of steam engines and air compressors; the cylinders 
and bearings of gas engines, kerosene engines, semi-Diesel engines, 
and Diesel engines ; the piston-rod glands of certain ammonia com- 
pressors; certain large and important bearings which for some 
reason or other must have the oil forced in under pressure to 
prevent wear; etc. 

In order to analyze the merits or demerits of the very numerous 
types of mechanically operated lubricators, some of the important 
features will be discussed as follows: 

Sight feeds. 

Pump plungers. 


Types of drive. 

Feed adjustment. 

Heating arrangement. 


Check valves. 



Sight Feeds. From this point of view, mechanically operated 
lubricators may be classified as follows : 

1. Those without sight feeds. 

2. Those with sight feeds on the suction side of the pumps. 

3. Those with sight feeds on the discharge side of the pumps. 
Mechanically Operated Lubricators without Sight Feeds. The 

Mollerup (so called after the inventor, a Danish engineer) me- 
chanically operated lubricator is the most widely used lubricator 
of this type in Europe. A large-diameter plunger is slowly forced 
into a cylinder filled with oil by means of a ratchet actuating 
motion combined with a worm-gear drive. The oil thus driven 
out is passed through piping to the engine. 

When the lubricator is being filled, air may be drawn into the 
cylinder, so that the lubricator does not start feeding immediately 
the engine starts, and lubrication difficulties may therefore arise 
before the lubricator starts to discharge the oil. 

Owing to the absence of sight feeds, irregular working of these 
lubricators, such as leakage past the pump plunger, is not always 
observed in time to prevent trouble. 

Some American mechanically operated lubricators have oil 
blinkers in the discharge line which act as the equivalent of 
sight feeds; they blink every time that oil is forced through but 
do not indicate the actual amount of oil passing. 

Other makers put two-way test cocks in the delivery pipes. 
When the handle of these cocks is turned to a horizontal position, 
the oil is delivered out through a test pipe into the atmosphere 
under no pressure; it is assumed that when the handle is turned 
vertical, the same amount of oil will be fed to the engine against 
pressure. If, however, the pump is not efficient, or if it is out of 
order, this will not be the case; less oil will be forced to the engine 
than is indicated by the test cock. 

In a multiple-feed, mechanically operated lubricator of this 
type, if one feed is choked and the others are working normally, 
it is impossible to locate the defective pump until the part of the 
engine supplied gives clear evidence of the lack of lubrication. 

Mechanically Operated Lubricators with Sight Feeds on the Suction 
Side of the Pumps. Some mechanically operated lubricators of 
this type (Fig. 7) have a container from which the oil is fed 
by gravity through sight feeds; the oil feeds are controlled 
by adjustable needle valves, and whatever oil drops into 



the pumps is forced to the engine, less possible leakage past 
the plungers. 

The disadvantage of these lubricators is that the oil feeds 
irregularly, because of variation in oil level and oil temperature. 
Furthermore, dirt is liable to choke up the needle valves and cause 
erratic oil supply. As the oil feeds are started and stopped by 
hand, these lubricators are not entirely automatic in action. 

1 Driving Cam 

3 Driving Rocker 
> Pump Plunger 

4 Bight Feed 

Oil Discharge 

FIG. 7. Mechanically operated lubricator with gravity sight feeds. 

Other mechanically operated lubricators, although they have 
the sight feeds on the suction side of the pumps, are fully auto- 
matic in action, the oil feeds starting and stopping with the 
engine. One type of these lubricators (Fig. 8) has a single 
plunger which on the suction stroke draws oil through a sight 
feed glass filled with water; on the delivery stroke the suction 
valve closes, and oil is forced out through a spring-loaded delivery 
valve. One important drawback to this arrangement is that the 
water in the sight-feed glass gradually disappears and is replaced 



by oil; this occurs even if a suction valve be placed below the glass, 
as it cannot be spring loaded; the author can see no virtue in 
the sight-feed glass's not being under pressure. Sight-feed glasses 

FIG. 8. Mechanically operated lubricator, sight-feed glass on suction side. 

seldom break because of internal pressure; they are either knocked 
to pieces, or they are fractured because of excessive strains set 
up when being placed in position. If the sight-feed glass is 


broken, the oil feed stops, as air is sucked into the sight feed in 
place of oil. 

All water-filled sight-feed glasses are liable to be fractured in 
the cold, if the water freezes. This is prevented by adding 
ordinary salt or glycerin to the water. 

With most lubricators, which have the sight-feed arrangement 
on the suction side of the pumps, one cannot be certain that the 
true oil feed is shown. If the 'pump plunger leaks on the delivery 
stroke, some of the oil will leak back to the oil container; this can- 
not easily be observed, and if the leakage is appreciable, it means 
that more oil passes through the sight feed than is actually dis- 
charged by the pump to the engine. 

Some lubricators have "dummy sight feeds. " One plunger 
pumps the oil through a sight feed, while a similar plunger 
pumps what is believed and hoped to be a similar amount of oil 
to the engine; the oil drops through the sight feed back to the 
oil container. Cases have occurred where one plunger was pump- 
ing oil merrily through the sight feed while the corresponding 
plunger was air locked. Strange to say, thousands of such lubri- 
cators have been sold, and engineers have not even taken the 
trouble to ascertain whether the sight feeds were true sight feeds 
or merely dummies. 

Several types of lubricators have two plungers for each oil 
feed. A measuring pump draws the oil from the container and 
discharges it under low pressure through a sight feed, whence 
it is sucked into the delivery-pump chamber and discharged 
through a check valve to the engine. 

Instead of two separate plungers, a two-diameter plunger is 
sometimes used, the small-diameter part acting as the discharge 
plunger. If there be any leakage from the discharge plunger, the 
oil can generally be seen filling up in the sight feed, and steps 
can be taken to rectify the trouble. With a two-diameter plunger 
properly constructed, all the oil passing through the sight 
feed is forced to the engine never less, as with leaky single 

Mechanically Operated Lubricators with Sight Feeds on the 
Discharge Side of the Pumps. Sight feeds that show the oil in 
the form of drops rising through water are true sight feeds, as 
they show the oil after it has left the pump and is actually on 
its way to the engine ; it cannot go anywhere else. 



Figures 9 and 10 show a cylindrical sight-feed glass, and Fig. 
11 a single bull's-eye sight-feed arrangement; the former sight 
feed will stand 300 to 400 and the latter 800 to 1,000 Ib. per 
square inch quite safely, when well made. 

The glass in Figs. 9 and 10 has both ends rounded and ground 
by a circular grinder, so that there are no sharp edges, whence 
fractures might emanate. 

FIGS. 9-10. Sight-feed glass under pressure. 

In a dark engine room it may be difficult to see the oil food in 
the bull's-eye shown in Fig. 11, so a better arrangement is to have 
a double bull's-eye with glasses both front and back. To keep 
the oil drops away from the glass it is good practice to have a 
climbing wire inserted in the nozzle from above (Fig. 11); the oil 
drops form, move up the wire, and unite with the oil at the top 
without removing any water; when there is no wire (Fig. 10), 
the drops wobble up through the water and usually lean against 
a corner, each drop enclosing and carrying away with it a small 
globule of water, so that the glass soon fills up with oil; this is 
avoided by having a climbing wire, as shown in Fig. 9. 


Another useful feature is shown in the shape of the nozzle 
(Fig. 9), this being narrow below the head. This prevents oil 
drops from sagging and creeping down the side of the nozzle 
and smearing the sight-feed glass, as in Fig. 10. 

A third point of importance for keeping the water in the glass 
is a spring-loaded check valve below the nozzle ; if this valve is not 
loaded, it "floats" after the delivery stroke has been completed; 
and if it is not seated at the beginning of the suction stroke, a 
little water may be sucked into the mouth of the nozzle; the result 
is that the glass slowly fills with oil. 

In very cold weather, steam-cylinder oil becomes very sluggish ; 
the oil drops become bigger, and even with a climbing wire, etc., 
the drops are inclined to take " pinpricks " of water away with 
them and slowly empty the glasses of water. 

If the pump is a good one and will pump water, a simple way 
of driving out accumulated oil from a sight glass and replacing 
it with water is to pour a small quantity of water into the lubri- 
cator container gradually, until the water begins to make its 
appearance at the sight-feed nipple in place of oil. Then add a 
little more, say an eggcupful or what seems necessary, and the 
water will be pumped up by the action of the lubricator, refilling 
the glass and driving the oil out. If the engine can be stopped, 
the proper method is to uncouple and clean the glass and fill up 
in the usual way. The method described is, however, useful 
where an engine runs continuously. 

Many engineers appear to be under the impression that a 
mechanical lubricator pumps oil only when a drop rises in the 
sight-feed glass. This is, of course, erroneous. Let us assume 
that it takes 1C strokes of the pump for one drop to rise through 
the glass; then for every stroke of the pump, the drop forming 
on the nozzle grows in size with a quantity equal to one-tenth 
drop; but as the glass is full of water, and the oil pipe leading 
to the engine completely filled with oil right to the check valve 
fitted at its extreme end, it must be clear that for every stroke 
of the pump one-tenth drop is forced into the sight glass at 
the nozzle and one-tenth drop is simultaneously discharged at 
the other end through the check valve. When the pump has 
made 10 strokes, the drop of oil formed on the nozzle becomes 
sufficiently large to overcome by its floating power its adhesion 
to the nozzle; the drop then rises, which simply means that 



it changes its position in the sight-feed glass, moving from 
the nozzle up to the top of the glass; this movement does not 
in any way affect the discharge of oil from the check-valve end 
of the oil pipe, which continues to be one-tenth drop every time 
the pump plunger completes its delivery stroke. 

Pump Plungers. These should not be too large in diameter, 
as then the pump strokes have to be very short and easily become 
irregular. Two-diameter plungers are advantageous, as the 

Fio. 11. Mechanically operated lubricator with bull's-eye flight feed. 

difference between the two diameters (see Fig. 11) can be made 
very small, say 3^4 in. (^ X l %4> in.); if the plungers have to 
operate at high speed and must supply only a small amount of 
oil (e.g., Diesel-engine cylinders), the stroke will still be percepti- 
ble, whereas with single plungers, }/ in. diameter, it would be 
well-nigh impossible to adjust the stroke to the required length 
and maintain it with certainty. 

Pump plungers should preferably not operate vertically 
with the oil below them, as they then easily become air locked, 


and it is difficult to let the air out. Plungers should either 
operate horizontally or, if vertical, should have the oil above the 
plunger discharge end. 

Outside plungers with packings should be avoided, as, if 
the plungers get scored, the leakage is difficult to overcome. 
It is better to have plungers inside the oil container and sealed 
by the oil; if they are hardened and ground to a good sliding fit, 
they will pump against considerable pressure with no or only 
slight leakage. 

It is bad practice to have two horizontal plungers operating 
together on opposite sides of the container and firmly connected ; 
it means that when they are a good fit, it takes great force to 
move them, as it is impossible to drill the pump cylinders in 
perfect alignment. Such a plunger arrangement causes exces- 
sive strain and wear of the driving mechanism. 

Valves. Most lubricators have single suction and delivery 
valves. If a valve becomes inactive by a piece of dirt's getting 
on to the valve seat, the lubricator may stop feeding. The 
author strongly recommends two suction and two delivery valves, 
so that one valve will act while the other is given a chance to 
get free of the dirt. The second delivery valve should be spring 
loaded to secure prompt closing. Spring-loaded suction valves 
are unsatisfactory, as the springs have to be very weak indeed, 
if they are not to interfere with the pump action on the suction 

The valves should be easily accessible the suction valves in 
particular. Figure 11 shows one method of placing the suction 
valves in a detachable cage. The pump should preferably 
be capable of freeing itself from air. With a spring-loaded 
delivery valve it becomes necessary to let the air out, in case of an 
air lock; this may be done, as shown in Fig. 11, by having a 
small air vent between the two delivery valves. This is opened, 
until all air is driven out, and oil appears at the vent; it is then 
closed, and the oil, having already passed the bottom valve, will 
force open the top valve. 

Some pumps do not have suction valves but suctipn ports, 
which are uncovered and closed by the movement of the plunger. 
A complete vacuum is created on the suction stroke, and, when 
the suction port is uncovered, oil is sucked in; but with viscous 
oils like steam-cylinder oils, the pump motion must be very slow, 



to ensure that the pump draws in a full charge of oil. A few 
lubricators have no valves at all but control the oil inlets and 
outlets by plungers very much like a piston-valve arrangement 
in steam engines; this arrangement requires most excellent and 
accurate workmanship to give satisfaction for high-pressure con- 
ditions. Whatever the valve arrangement may be, it is always 
desirable that the suction passages be as short and wide as 
possible (to avoid wiredrawing of the oil) and that the plungers 
operate with small pump-chamber clearance. 

Types of Drive. The principal methods of driving mechanical 
lubricators arp 

Lever drive. 

Rotary drive. 

Worm-gear drive. 

Spur-gear drive. 

Ratchet drive and ball or roller-clutch drive. 

Lever Drive (Fig. 12). The plunger is operated by a rocker, 
which gets its motion from some part of the engine, e.g., the 
half-time shaft on a gas engine 
(Fig. 171, page 465). In this 
way the movement of the 
plunger can be made to syn- 
chronize with the piston move- 
ment, and the oil injected at a 
definite movement in the cycle. 

In large, slow-speed, long- 
stroke steam pumping engines, 
the oil can in this way be forced 
into the steam just at the 
moment when it is required. It 
must be noted, however, that 
such timed injection of the oil 
can take place only in lubrica- 
tors that pump oil alone and not oil and air, as do most lubri- 
cators in which oil drops through a sight feed into the delivery 
pump. If air gets pumped into the oil pipes, it has a cushion 
ing effect, and oil is discharged only when the back pressure is 
at its minimum. 

Rotary Drive. The lubricator shaft has a driving pulley out- 
side the container; the shaft revolves and may, by means of a 

~O O Q O 

[,, :>:;:, '>Zf& 

ybo o o o 

Fio. 12. Lever drive. 



cam, actuate the plunger. Obviously, this form of drive can 
in this way be adapted to time the injection of oil from the 
various plungers, by suitably spacing the cams on the lubricator 

In most lubricators the cams do not actuate the plungers direct, 
as in Fig. 11, but by some intermediary mechanism, which in the 
majority of cases is rather unmechanical. The most common 
form is that of a cam revolving eccentrically between two jaws 

Fio. 13. Cam motions. 

or inside a slot, as indicated in Fig. 13, but a cylindrical surface 
does not wear well with a flat surface; the result is therefore more 
or less rapid wear; such motions fairly soon develop considerable 
backlash, which increases the wear. Figure 13 shows one 
method of preventing wear with a cam drive; the cam has a loose 
roller, which, when pressed against the plunger head by the cam, 
remains stationary during the delivery stroke; the cam revolves 
inside the roller, and, as it is well lubricated, there is no wear 
whatever or any side pressure on the plunger. 


Worm-gear and spur-gear drives are used for operating lubri- 
cators on high-speed engines or machinery, so that the pump 
plungers may be made to operate at a comfortable speed and with 
fairly long strokes. 

Ratchet drive and ball or roller-clutch drive is used when the 
motion is taken from some reciprocating part of the engine, e.g., 
one of the valve rods on a steam engine (see Fig. 14). Ratchet 
drive is usually preferable to clutch drives, except at low speeds, 

01 LOl [01 [01 [0 

01 101 [01 [01 [O 

FIG. 14. Ratchet-drive arrangement. 

when there may not be much to choose between them. At high 
speeds, the balls and rollers in clutches wear out the casings, and 
slipping begins, with the too frequent result that the lubricator 
stops working. 

High-speed ratchet drives must be carefully designed; the 
ratchet wheel should be made of casehardened tool steel and 
screwed on to the shaft in such a manner that the motion tends 
to keep it in place. The driving as well as the backlash pawls 
should be made very light, preferably of thin folded steel plate, 
which presses only lightly against the teeth in the ratchet wheel ; 


heavy pawls, due to inertia forces, do not act promptly, unless 
backed by powerful springs, in which case rapid wear takes place. 
The ratchet should be rather small and should not move more 
than two or three teeth per stroke; otherwise, the driving pawl 
will strike the teeth too hard. Occasionally, a ratchet wheel 
will jump forward several teeth, owing to lack of resistance; this 
occurs chiefly when the lubricator has only one or two plungers 
to operate and can be overcome by tightening the gland packing 
on the lubricator shaft, where it passes through the container, or 
by fitting some sort of brake on the shaft. 

Lubricators for exposed conditions, e.g., locomotive lubricators, 
should have the ratchet wheel enclosed in an oiltight casing filled 
with oil, or the ratchet should bo inside the container. 

Feed Adjustment. With ratchet drive an alteration in feed 
is made by altering the leverage or angular movement of the 
actuating arm, which moans a greater or smaller number of 
strokes per minute. An alteration in the amount of oil fed per 
stroke can be made by having a by-pass on the delivery side, 
by wiredrawing the oil inlet (suction passage), by altering the 
stroke of the plunger, by keeping the suction valve or port open 
for part of the delivery stroke, etc. 

The first two methods are very unsatisfactory, particularly 
with viscous oils, as any alteration in viscosity means an altera- 
tion in oil feed. One method of altering the plunger stroke is 
shown in Fig. 11, viz., by altering the position of the two adjust- 
ing nuts; they may be so adjusted that the plunger is never 
touched by the cam roller no-stroke position or they may 
allow the cam roller to touch the plunger all the time full-stroke 
position any intermediary position can also be secured. 

Keeping the suction ports or valves open during part of the 
delivery stroke has the same effect as shortening the plunger 
stroke and, with a well-designed arrangement, is capable of giving 
good results. With the two last-mentioned methods the oil 
feed, assuming that the valve arrangement is satisfactory, will 
be maintained uniform and independent of the viscosity of the 
oil, as long as the speed is low enough and the oil fluid enough 
at the working temperature entirely to fill the pump space on 
the suction stroke. With steam-cylinder oils the number of 
long strokes per minute must not exceed 20 to 30 to get perfect 
pump action, say above 90 per cent volumetric efficiency; with 



medium-viscosity internal-combustion engine oils, a speed of 
250 to 300 short strokes per minute may be permitted. 

There are multiple-feed lubricators in which one large master 
pump supplies oil for a number of delivery pumps, the feed to 
each of them being controlled by a drip-sight feed; the surplus 
oil delivered by the large pump over and above what is taken by 
the delivery pumps is by-passed back to the container through 
a loaded check valve. In this arrangement the oil feeds are 
much influenced by alteration in viscosity of the oil (temperature 
changes) ; also, an alteration in one of the feeds affects the other 

For these reasons the author is a strong advocate of separate, 
independent, and interchangeable pump units for each oil food, 
e.g., the pump unit in Fig. 11, which represents a 
design by A. Kirkham and the author. But this 
principle of soparatc pump units for each food can, 
of course, bo appliod to any number of designs. 

Heating Arrangement. Lubricators that aro 
oxposod to low temperatures and have to pump 
viscous oils, e.g., lubricators on locomotivos, steam 
traction engines, etc., must bo fitted with heating 
tubes. Usually, a straight tubo through tho con- 
tainor as near the suction ports as possible, or 
ovon a short hollow tube scrowod into the con- 
tainer, will provo adequate; they must bo con- 
nected to the steam supply, say 10 min. before 
starting, so as to liquefy the oil sufficiently to 
ensure good pump action. 

Strainer. Most lubricators have a shallow 
perforated strainer through which viscous stoam- 
cylinder oil passes so slowly that tho avorago 
driver never troubles to use tho strainor but takos 
it out when he fills the lubricator; ovon if it is not 
removed, it retains only coarse impurities. It is 
best made of gauze, which has finer openings than 
perforated plate and yet a considerably greater area of openings 
to pass the oil. The strainer should be deep, as shown in Fig. 11, 
and with a solid bottom and rim, so that any dirt or water in the 
oil may accumulate here while the oil filters through the 

Fio. 15. Chock 


Check Valves. At the extreme end of the oil pipes should be 
fitted spring-loaded nonreturn valves to prevent the oil pipes 
from emptying themselves; the force of the spring should be 
20 to 25 Ib. per square inch, so as to prevent a vacuum from open- 
ing the valve and sucking oil out of the pipe and lubricator; this 
is not an unusual occurrence with badly made check valves. To 
ensure good seating of the valve, the author favors ball valves 
with the spring soldered on to the ball ; this prevents the ball from 
rotating, and it forms a good permanent seating which should 
preferably be very narrow. 

Figure 15 illustrates one type of check valve which the author 
has used with great success. Figure 155 (page 418) shows a 
locomotive-pattern check valve. 

Desirable Features in Mechanically Operated Lubricators. In 
the author's opinion the things to aim at in the manufacture of a 
first-class mechanically operated lubricator are the following: 

1. Oil feeds independent of each other. 

2. Oil feeds independent of viscosity, oil level, or back pressure. 

3. Sight feeds showing the correct amount of oil actually passing out from 
the lubricator. 

4. Oil feeds capable of quick adjustment between wide limits. 

5. Freedom from air lock. 
(>. All adjustments outside. 

7. All partis easily accessible for adjustment, examination, or cleaning. 

8. No joints under pressure except final discharge. 

9. Low wear of parts. 

10. Efficient strainer. 

11. All pump units made up of standard, interchangeable parts. 

12. Adaptability for ratchet drive, rotary drive, worm-gear drive, spur- 
gear drive, or oscillating-lever drive. 

13. Simplicity and compactness of design. 

14. Low cost of manufacture. 


(Bearings in General) 

Bearings are used to support the revolving or oscillating parts 
of engines and machinery, and the problem of bearing lubrication 
is therefore the oldest of all lubricating problems. 

In the early days, bearings were crudely designed, and low- 
speed conditions prevailed. The lubricating mediums were vege- 
table oils, such as olive oil; rapeseed and castor oil; and animal 
fats and oils, such as tallow, lard oil, sperm oil, and whale oil. 

The enormous development of modern engines and machinery 
has brought into existence a variety of bearings operating under 
higher speeds, higher pressures, or higher temperatures than at 
any time before. Lubricating oils to suit modern conditions have* 
of necessity undergone a similar great development, made possible 
by the production of mineral lubricating oils manufactured from 
a variety of petroleum crudes. The subject of bearing lubrica- 
tion will be divided into several sections as follows: 

Bearing Materials. 
Operating Conditions. 
Oiling Systems. 
Frictional Heat. 
Bearing Troubles. 
Selection of Oil. 
Bearing Oils. 
Semisolid Lubricants. 
Solid Lubricants. 


Bearings are made in all sizes from very small to very large, and 
there are two main types, as follows: 



Journal Bearings Thrust Bearings 

1. Solid bearings. 1. Plain thrust bearings. 

2. Two-part bearings. 2. Ball and roller thrust bearings. 

3. Four-part bearings. 3. Michell and Nomy bearings. 

4. Ball and roller bearings. 

5. Michell and Norny bearings. 

Journal Bearings. 1. Solid Bearings. Horizontal solid bear- 
ings are always small in size, used as inexpensive bearings for 
loose pulleys and small shafts and in a variety of machinery where 
slow speeds or low bearing pressures prevail or where the lubri- 
cating conditions are so excellent that little or no wear is antici- 
pated. This type of bearing is used as gudgeon or wrist-pin 
bearing in the great majority of high-speed steam engines and 
internal-combustion engines. 

When more than slight wear is likely to take place a bushing is 
frequently provided so that when the bushing is worn it can be 

Vertical solid bearings are used as neck bearings arid footstep 
bearings for high-speed spindles in textile mills, also as footsteps 
for vertical shafts. 

2. Two-part Bearings. The majority of bearings are of this 
type. For shafting bearings the two bearing halves are usually 
of cast iron, and the bearing comparatively long. They may bo 
hand oiled, drop-feed oiled, or arranged for ring oiling. 

In larger journals, bearing brasses are fixed in the top and 
bottom part of the bearing, and between the top and bottom 
brasses are placed "liners," which are thin strips of metal. When 
the bearing wears, one or more of these strips may be removed, 
so as to bring the two bearing brasses closer together around the 
shaft. Two-part bearings are often lined with antifriction metal. 

When the pressure is always taken by one of the brasses, say the 
lower one, as in many bearings, the top half of the bearing need 
not be very strong, nor does the top brass need to fit the journal 
closely; in many cases the top half then simply acts as a dust 
cover and to hold the lubricator. Railway axle boxes, for 
example, have only a top brass, the pressure being directed 
upward, and below the journal is a cellar, holding a pad oiler or 
waste packing for the purpose of lubricating the journal. 

A two-part bearing is not suitable where the pressure from the 
journal is directed against the joint of the two bearing halves; 


large bearings operating under such conditions are therefore 
frequently designed as four-part bearings. 

3. Four-part Bearings. These bearings are used principally 
as main bearings in large horizontal steam engines and gas engines. 
The bearing surface is built up of four parts, i.e., a top and bottom 
brass and two side brasses. 

4. Ball and roller bearings are described in a special chapter. 

5. Michell and Nomy bearings are described in a special 

Thrust bearings are designed to take up pressure in the direc- 
tion of the shaft,, the propeller thrust in the case of marine 
steam engines and turbines. A special chapter is devoted to the 
description of plain thrust bearings, ball and roller thrust bear- 
ings are described under ball and roller bearings, and Michell 
and Nomy thrust bearings are described in the special chapter 
devoted to these bearings. 


With perfect oil-film lubrication the nature of the rubbing 
surfaces does not influence lubrication, but most bearings are 
imperfectly lubricated; they wear more or less, and the various 
bearing metals behave differently. 

Bearings are chiefly metals, but wood, rawhide, fiber, agate, 
and jewels are used for special purposes. 

Bearing Metals. The journal and the bearing should prefer- 
ably be of dissimilar materials to work well together, and the 
bearing surface is usually of a softer material than the journal. 
If wear takes place, it will then be chiefly on the bearing surface, 
which is cheaper to replace than the journal. 

Good bearing metals must possess the following properties: 

1. Sufficient strength to sustain the load. 

2. Low running temperature, which means high thermal conductivity; 
white metals containing a high percentage of lead are inferior in this respect 
to those rich in tin and containing little or no lead. 

3. Low Coefficient of Friction. Hard bearing materials, such as the rigid 
bronzes (copper-tin alloys low in lead), are best in this respect, assuming 
that the bearing surfaces are carefully fitted to the journal; otherwise, white 
metals give lower friction, as they yield slightly and distribute the load more 

4. Durability. The rigid bronzes, and alloys containing zinc, wear more 
than those alloys which are rich in lead, but the latter have a higher coeffi- 


cient of friction. According to Dudley, those bearing metals will wear 
the least which have a fine granular structure and combine great elongation 
with great tensile strength, the elongation, however, being the more impor- 
tant property of these two. 

6. Low Journal Wear. The white metals excel over other metals. 

6. Ease of Replacement. Again here the advantage lies with the white 

7. Resistance to Corrosion. Tin and antimony resist corrosion best; iron, 
copper, lead, and zinc are more easily corroded, particularly the two latter. 
When the oil is likely to contain a large amount of free fatty acid, the white 
metal should preferably contain no lead and little or no zinc. 

The following combinations of bearing metals represent current practice: 

Hardened Crucible Steel on Steel or Bronze. For high pressure and low or 
moderate speed, e.g., hard-steel toggles working against mild-steel seats in 
stone breakers, presses, etc. 

Mild Steel on Bronze. For moderate pressures and low or moderate 
speeds, as exist in many important bearings. 

Mild Steel on White Metal. For low or moderate pressures and moderate 
or high speeds. This is the combination used in the great majority of 
machinery bearings. 

Mild Steel on Cast Iron. For low or moderate pressures and low speeds, 
as in textile machinery and the like; also used for small or medium-size 
shafting bearings; the bearings are long, and the pressures low; with higher 
bearing pressures, the cast iron must be lined with white metal. 

Cast Iron on Cast Iron. For low pressure, chiefly used for piston rings, 
cylinders, crossheads, and crosshead guides in steam engines and internal- 
combustion engines. 

Hard Steel, Bronze, or Brass. With all hard bearing metals it 
is important that the bearing surfaces be well scraped together 
with the journal and that the bearings be carefully erected, so 
that the pressures will be evenly distributed over the entire 
bearing surfaces; otherwise, certain parts of the bearings will 
be excessively loaded and cause heating. 

White metals (antifriction metals) are combinations of hard 
metal, such as antimony, embedded in a soft plastic groundmass, 
such as lead. 

When lined with suitable antifriction metal, which has more 
or less resilience, the journal easily beds itself down and distributes 
the pressure uniformly over the entire bearing surface. 

In high-speed steam and internal-combustion engines, where 
three, four, or five bearings support the crankshaft, the bearings 
are nearly always lined with white metal, with a view to distribut- 
ing the load equally over them all. 


If bronze is used and if, say, one bearing is slightly out of line, 
the bronze, not yielding, will create excessive bearing pressure in 
that particular bearing and cause heating. 

It is the hard grains in a white-metal surface that sustain 
the load; if the load is excessive at any point, the plastic body of 
the metal will yield until the load is evenly distributed over a 
great many hard grains; this will assist the lubricating oil in 
maintaining a good film everywhere and means increased safety 
in operation. If there are only a few hard grains in a white 
metal, it will be soft and will stand only low bearing pressures; 
if there are too many hard grains, the points of the hard crystals 
will engage one another and form a solid network throughout 
the body of the metal, which will then be found to be brittle. 
Trimetal alloys appear to give better service than those white 
metals which are composed of only two metals. 

Cast iron is porous and granular in structure; close-grained 
cast iron is best and can be obtained harder or softer as required. 
It is capable of attaining a very smooth, hard, and glazed surface; 
but if this surface is cut it takes considerable time to reproduce 
thfc hard glossy skin so very desirable from a lubrication point 
of view. 

Cast iron when not exposed to undue pressure and when well 
lubricated is a very satisfactory bearing metal. 

The use of graphite in connection with cast iron is capable of 
giving excellent results, as mentioned under "Solid Lubricants." 

Wood, Rawhide, and Fiber. Hard and dense wood is used to 
some extent for spur and bevel gearing in windmills, water mills, 
etc. For certain bearings, such as footsteps for water turbines 
and stern tube bearings, lignum vitae is favored, as it will stand 
great pressure, is of a greasy nature, is not easily abraded, and 
works well with water. 

Rawhide and fiber, also compressed paper, are sometimes used 
for pinion wheels and give silent running. 

Agate and Jewels. In watches and light machinery, which 
cannot be regularly lubricated, agate and various jewels are 
used as bearings for hard-steel pins. 


Workmanship may be defined as the attention that has been 
given to 


1. The finish of the bearing surfaces. 

2. The bearing clearance. 

3. The alignment of the erected bearing. 

Finish of Bearing Surfaces. The rubbing surfaces are never 
exactly true and smooth. If a new shaft is put into new bearings 
without oil, it will, when revolving, touch the bearing surfaces 
only at certain points, distributed more or less evenly over 
the surface. It is for this reason that bearings are " scraped 
together. " It is an advantage to have the surface of the shaft 
made as smooth as possible, and the high points in the bearing 
surfaces are scraped down until finally the shaft bears uniformly 
on the whole of the bearing area. 

Bearing Clearance. The diameter of the shaft is slightly 
smaller than the inside diameter of the bearing. The difference 
between the two diameters the bearing clearance should be 
about 1/1,000 in. per inch diameter of the shaft rather more 
than this for small bearings and rather less for large bearings. 

When the bearing surfaces are well lubricated, and particularly 
when they are supplied with a continuous stream of oil, which 
carries away the frictional heat, the bearing clearances can*be 
made smaller, and more efficient lubrication can be obtained 
than where bearings are boundary lubricated and the journals 
therefore are more likely to heat and expand. 

Alignment. When machinery and shafting are erected, it is 
very important that the various bearings be truly and accurately 
fitted. If, for instance, a length of shafting is supported by a 
number of bearings, and some bearings are placed too high and 
others too low, this will set up stresses in the shafts and in the 
bearings, creating difficult lubricating conditions. 


Size of bearing (diameter). 
Speed of shaft (surface speed per minute). 
Bearing pressure (pounds per square inch). 
Bearing temperature (degrees Fahrenheit). 
Mechanical conditions (good or bad). 

Size of Bearing. Tlvc surface of the shaft or journal is never 
perfectly smooth or round but will possess a roughness which, 
if not visible to the naked eye, can be seen through a magnifying 
glass. The imperfection in manufacture will have a tendency 
to produce metallic contact between the rubbing surfaces. This 


tendency is greater the larger the bearing, and experience has 
proved that, other things being equal, the larger the bearing the 
heavier in body must be the oil to provide efficient lubrication. 

Speed of Shaft. A revolving shaft will draw the oil into the 
bearing owing to the oils adhering and clinging to the shaft. 
Speaking generally, this action increases with the speed of the 
shaft and the body of the oil. When bearings operate at low 
speed, the oil used must bo heavy in body, and groaso may be 
preferable in some cases. At higher speeds, an oil light in body 
should preferably bo used; and for very high speeds, oils very 
light in body must bo used. 

Sommerfeld gives the following formula for the minimum spood 
at which the friction in the bearing changes over from " boundary 
friction" to fluid friction: 

v . , I...H 

" llfl * 151 

where V is expressed in motors por second. 

P is bearing pressure in kilograms por square centimeter. 
8 is bearing clearance in centimeters. 
r is radius of shaft in centimeters. 
17 is absolute viscosity in poises. 

At extremely high speeds, air even has boon used as the only 
lubricant, as in the case of spindlo bearings for traverse spindle 
grinders used in watch factories. The spindles aro }^ in. in 
diameter. Both spindles and bearings aro of hardened stool and 
fitted together with extreme oaro; the fit is so close that when 
thoy aro not running it is difficult to slide the spindlo through the 

When starting up, the spindles will give a grating noiso for a 
few seconds; but when attaining their normal spood of about 
12,000 r.p.m., thoy run quite smoothly and with so little friction 
that, when the driving belt is thrown off, they continue to run for 
a couple of minutes until the air film breaks and the spindlos 
quickly stop. The surfaces must be kept vory clean by rubbing 
with alcohol and tissue paper. If the bearings or spindles aro 
not perfect, a little kerosene needs to bo used to give smooth 

Bearing Pressure. Bearing pressures rango from a fow pounds 
per square inch for cast-iron piston rings rubbing against cast- 


iron cylinders to as much as 3,000 to 4,000 Ib. per square inch for 
hardened steel rubbing against steel, as in slow-speed punching 
machines. The bearing pressures are chiefly governed by the 
nature of the bearing materials, the character of the load, and the 
degree of lubrication efficiency desired. 

For ordinary conditions the bearing pressures permissible for 
various metals are indicated in the following table: 

Pressures, Pounds 
per Square Inch 

Hardened crucible steel on steel 2,000 

Hardened crucible steel on bronze 1 ,200 

Unhardened crucible steel on bronze 800 

Mild steel with smooth compact surface on bronze. . . 500 

Mild steel with ordinary surface on bronze 400 

Mild steel with ordinary surface on white metal 500 

Mild steel on cast iron 300 

Cast iron on cast iron (journal bearings) 100 

These figures may be increased 50 per cent, 100 per cent, or 
even more, if the load is intermittent, also if the bearings are 
well cooled, as in locomotive crankpins and crossheads. 

The figures must be reduced if the pressure is always in one 
direction and never relieved; also, if it is important that no wear 
should occur, as in many electrical machines and other high-speed 
engines, such as enclosed-type steam engines and gas engines, 
lubricated by a circulation oiling system. If wear must not 
take place, it means that the bearings must have perfect oil-film 
lubrication at all times; with high surface speed, higher bearing 
pressures may be allowed, as the oil film is more easily formed. 
But the accuracy and smoothness of the working surfaces is also 
exceedingly important, to enable an unbroken oil film to form. 
This intricate subject has been treated mathematically in a 
very thorough and practical manner by E. Falz, in his book 
"Grundzuge der Schmiertechnik." 1 

All other things being equal, it is obvious that the greater the 
pressure on the bearing and the lower the speed the heavier in 
body must the oil be to sustain the pressure without being 
squeezed out too rapidly. If the pressure on the bearing is 
slight, light-bodied oil can be used, and at moderate or high speeds 
a moderate oil supply will be sufficient to maintain a complete 

1 Julius Springer, Berlin, 


oil film. If the pressure on the bearing is great, an oil heavy in 
body must be used; and if, in addition, the speed is low, it is very 
difficult, if not impossible, to maintain a complete oil film and to 
prevent metallic contact of the rubbing surfaces. Under such 
conditions certain solid or semisolid lubricants may prove more 
efficient than lubricating oils. 

Bearing Temperature. Where machinery is operating in cold 
surroundings, or at very low speeds, bearing temperatures may 
be low (from 70 to 90F.). When bearings operate in very cold 
surroundings, light-bodied oils and oils with low cold tests should 
be employed, so as not to congeal and cause difficulty in feeding. 

The majority of bearings operate at medium temperatures, 
from 90 to 120F. High-speed bearings frequently operate at 
temperatures higher than 120 but seldom above 160F. Bearing 
temperatures above 120F. must be termed high and should 
ordinarily never be allowed to exceed 140F. (see "Turbines"). 

If the bearing temperature is higher than 160F., the conditions 
should be carefully looked into, as such temperatures are dan- 
gerous and show that the mechanical conditions are wrong 
and should be corrected or that the quality of the oil used is 
unsuitable or that an insufficient quantity of oil reaches the 
parts to be lubricated. 

If bearing temperatures are high, notwithstanding that the 
mechanical conditions are correct and that carefully selected 
good-quality oil is used in sufficient quantity, the conditions are 
evidently so severe that the heat developed in the bearing cannot 
be radiated quickly enough from the bearing surface. In sueh 
cases a circulation oiling system should be introduced, in order to 
remove the frictional heat and reduce the bearing temperature 
sufficiently for safe operation. 

Mechanical Conditions. Bearings, in time, will usually wear or 
get out of alignment; it is important that they be kept in good 
alignment and repair, by renewing bushings, brasses, or anti- 
friction linings; by adjusting bearings for wear; etc. 

When trouble or irregularity in operation occurs, the cause 
should be traced, and the conditions rectified, rather than that 
the trouble should be allowed to continue until it become** serious. 

By good mechanical conditions is understood bearings of good 
design; journals and bearing surfaces of good material, well 
finished and with suitable bearing clearance; bearings in good 


alignment and not appreciably worn; also reasonable atten- 
tion to regular oiling of the bearings. 

By bad mechanical conditions is understood bearings that are 
crudely designed, or of good design but allowed to get out of 
order; bearings made of poor or unsuitable material; bearing sur- 
faces rough or worn, bearings out of alignment ; also lack of atten- 
tion in keeping the oiling system in its most efficient state. 

Speaking generally, bad mechanical conditions necessitate the 
use of oils heavy in body, whereas under good mechanical condi- 
tions oils lighter in body can be employed, resulting in more 
efficient lubrication of the bearings. 


The various systems by which oil is applied to bearings may 
be divided into seven main groups, as follows: 

Individual bearings 

Hand oiling. 
Drop-feed oiling. 
Pad oiling. 

Ring oiling. 
Bath oiling. 

~ - , (Splash oiling. 

Groups of hearings -I * . .. ... 

/Circulation oiling. 

Oiling Systems for Individual Bearings. Hand oiling is the 
oldest system employed for lubricating bearings; it is the least 
efficient and the most wasteful of all oiling systems. Hand oil- 
ing is employed for lubrication of low-speed shafting and low- 
speed bearings in a variety of machines, such as machine tools, 
textile machinery, and printing machines. It is largely employed 
for oiling small parts of valve motions, valve spindles, etc., in 
steam engines, internal-combustion engines, and other power 
producers. It is also employed on various types of machines 
exposed to heavy vibration or rough use where any kind of 
lubricating appliance would be shaken off. 

In the bearing is a hole, usually in the top part. The oil is 
applied by an oilcan, preferably of the press-button type, by 
which it is possible to deliver one drop or a few drops of oil as 
required, in order not to waste it. The oil runs down the hole, 
spreads over the bearing surfaces, and gradually works its way 
toward and out through the ends of the bearing. After each 
oiling, the oil film in the bearing gradually becomes thinner, and 


finally the bearing runs practically without lubrication until such 
time as it is oiled afresh. 

The lubrication will gradually decrease to a state of inefficiency, 
dependent upon the body of the oil in use, the length of time 
between oilings, and the operating conditions. 

In order to prevent the entrance of dust or fluffy matter, which 
would tend to choke up the oil hole or would enter the bearing 
and cause trouble, the entrance to the oil hole may be fitted with 
an oil-hole protector (see Figs. 109, 110, 111, page 302). Another 
method is to provide a felt pad in the oil hole into which the oil 
is poured. This method insures more uniform feeding of the oil. 

In many cases the oil is not applied through an oil hole but 
simply to the end of the bearing, e.g., with textile machinery. 

Drop-feed Oiling. The drop-feed oiling system includes all 
appliances by which a moderate and more or less regular supply 
of oil is fed to the bearing. 

There are four types of such appliances, viz.: 

Siphon oiler. Sight-feed drop oiler. 

Bottle oiler. Mechanically operated lubricator. 

Siphon Oiler. When, in the early days of engineering, hand 
oiling proved inadequate for lubricating heavy-duty bearings, 
the siphon oiler was the first improvement introduced. It (Fig. 
16) consists of a container (1) in which oil is filled to a certain 
level; the siphon oil tube (2) projects above the oil level; the 
siphon wick is introduced into the oil tube, its lower end being 
at a lower level than the end immersed in the oil. The oil level 
should not be allowed to be higher than the top of the oil tube, 
as the surplus oil will then be wasted through the tube. 

With siphon oilers the oil feed varies with the oil level in thc3 
container, also with the temperature of the oil, as cold and thick 
oil will feed more slowly than warm arid thin oil. 

The siphon wick consists usually of one or several strands of 
woolen yarn, preferably of loose texture, which feed more freely 
than yarns of tight twist and close texture. The higher the oil 
level in the container, or the thinner the oil, or the deeper the 
siphon is introduced into the oil tube, or the greater the number 
of strands in the siphon the greater will be the oil feed. When 
so many strands are used that they choke the oil tube, a point is 
reached where the addition of more strands will reduce the oil 



feed because of the greater resistance in passing through the tight 
siphon; choke trimmings used in locomotives (Fig. 98, page 284) 
are of this type. 

The container should always be fitted with a lid, so as to pre- 
vent the entrance of dust, dirt, and water into the oil. Siphons 
in time get choked with impurities and become inoperative; 
they should be renewed at suitable intervals. 

Siphon oilers are rather wasteful but very reliable where a 
moderate oil feed is required; they are not suitable for very small 
feeds. Where machines or engines are running intermittently, 
the siphons should be lifted out of the oil tube and left in the oil 

FIG. 10. Siphon oiler. 

container every time that the machinery stops; otherwise, they 
keep on feeding, and oil is wasted; oil should be added to the con- 
tainer at frequent intervals so as to keep the oil level as constant 
as possible. 

Siphon oilers are employed for lubrication of locomotives, 
marine steam engines, main bearings of old-type stationary steam 
engines, and other prime movers, as well as for the lubrication of 
medium-size bearings of shafting and in a variety of machines of 
all kinds. 

The oil container may have several siphon tubes, each tube 
being served by a separate siphon; such multiple-feed siphon 
boxes are occasionally fitted with sight-feed glasses below the 
container, so that the oil feed from each siphon tube is visible. 




The bottle oiler (Fig. 17) has been specially developed for the 
lubrication of light and medium-sized shafting bearings operating 
at low to moderately high speed and under conditions that 
make a small constant feed desirable. The glass bottle (1) has a 
stopper (2) fitted with a brass tube (4). A copper or steel needle 
(3) fits loosely inside the brass tube, its lower 
end resting on the shaft in the bearing. 

The shaft, when revolving, gives the needle 
a very slight up-and-down motion, which has 
the effect of drawing a sparing supply of oil 
from the glass bottle, the oil creeping down 
over the surface of the needle and finally 
reaching the bearing surface. 

The bottle oiler is automatic in action, 
starting and stopping with the motion of the 
shaft. If the bearing gets warm, the needle 
heats up; the oil surrounding the needle 
becomes thinner, and more oil will be fed. 
If the bearing vibratos, the greater movements Fia. 17.- 
of the needle will result in more oil's being fed. 
If it be found that the amount of oil supplied through tho bottlo 
oiler is insufficient, the oil feed can be increased by using a 
thinner needle or by filing a flat on the side of the needle. 

The stopper should preferably have a brass tube, as shown, in 
which the needle has a loose sliding fit; without this tube tho 
opening in the stopper varies considerably and in time causes tho 
oil food to stop on account of tho swelling of the stopper. 

Bottle oilers cannot bo used on machinery exposed to rough 
use, as, being of glass, they are easily broken. 

The sight-feed drop oiler (Fig. 18) has largely replaced tho 
siphon oiler. It can be adjusted to food one drop of oil per minute 
or more. It consists of a container, usually having a glass body 
so that the level of tho oil can bo observed. The adjusting needle 
or valve spindle (5) is guided into a conical hole in the bottom of 
tho oiler. By turning the milled collar (3) the needle can bo 
raised or lowered so as to give a greater or smaller feed. If tho 
handle (4) of the top of the adjusting needle 5 is turned to its 
horizontal position, the needle drops by spring tension and shuts 
off the oil supply; when it is again raised, the feed will be the same 
as previously adjusted. 



The sight-feed drop oiler has the same disadvantage as the 
siphon oiler as regards variation in oil feed, due to higher or 
lower oil level or due to the oil being cold and thick or warm and 
thin; in addition, when adjusted to feed a very small amount of 
oil, grit and dirt may easily choke the oil outlet from the oiler, 
so that the feed stops altogether. The sight-feed drop oiler 
has the advantage over the siphon oiler in that the feed can be 
quickly adjusted, quickly started and stopped, and the oil level 
as well as the oil feed is clearly visible. 


Glass Body 

Oil Level 

Adjusting Collar 

Shut- off Handle 

Top of Adjusting Spindle 

Sight Feed 

Foot Valve 

(only used when feeding 
against intermittent pres- 

FIG. 18. Sight-feed drop oiler. 

Sight-food drop oilers may be arranged to havo more than one 
food. For example an oil container may have six oil outlets, 
controlled by six different needle valves, the oil dropping through 
sight foods into oil tubes which guide the oil to the different 

Sight-food drop oilers are extensively used on modern steam 
engines and power producers of all kinds. 

When feeding oil to the crankpins of steam engines, gas engines, 
and other prime movers, the so-called crankpin banjo oiler is 
of ton employed (see Fig. 171, page 465). 

The Nugent crankpin oiler, much used in the United States, is 
shown in Fig. 19. The sight-feed drop oiler is held in a vertical 
position by the weighted pendulum (1) to which it is attached. 




FIG. 19. Nugent crank- 
pin oiler. 

The part (2) revolves centrally, receives the oil through the tube 
(3), and guides it to the crankpin. 

A mechanically operated lubricator, either single feed or multi- 
ple feed, is occasionally employed for feeding oil to important 
bearings. The advantage is that, being 
operated from a moving part of the engine, 
the mechanically operated lubricator starts 
and stops with the engine and feeds the oil 
more uniformly and regularly, therefore 
with less waste, than when sight-feed drop 
oilers or siphon oilers are used ; also a much 
more viscous oil can be fed, if required. 
The various feed pipes are preferably fitted 
with check valves at their extreme ends in 
order to ensure that the pipes are always 
filled with oil, so that as soon as the engine 
starts, and therefore the lubricator, the 
oil will immediately be delivered from the 
ends of the oil pipes. 

Sight-feed arrangements are either fitted in the lubricator 
itself, one sight feed for each oil feed, or fitted at the extreme 
ends of the oil pipes, the oil dropping from the check valves 
through sight feeds into the bearings. 

Pad Oiling. Lubrication by pad oilers or oil-soaked waste is 
used chiefly in railway practice and described under " Railway 
Rolling Stock." . 

Ring Oiling. This method is very efficient and is described in 
a special chapter. 

Bath Oiling. This system is employed only for vertical bear- 
ings, such as bail bearings; high-speed bath spindles employed 
in textile mills; or the footsteps of vertical, heavy shafts, some- 
times found in textile mills, flour mills, vertical water turbines, 
vertical hydroextractors, gyratory crushers, etc. (see under 
respective headings). 

Oiling Systems for Groups of Bearings. Splash oiling is 
employed for lubricating a number of bearings enclosed in an 
oil tight casing, this system being frequently employed for 
lubricating enclosed vertical or horizontal steam engines, air 
compressors, gas engines, kerosene engines, gasoline engines, and 


The enclosed crank chamber is filled with oil to a certain level ; 
means should be provided to maintain this level as constant as 
possible. Dippers fixed to the crankpin bearings (big ends) dip 
into the oil and produce inside the crank chamber a spray of 
tiny drops of oil which reach and lubricate the main bearings, 
crankpins, gudgeon or wrist pins, cams, and various other bear- 
ings or parts. The bearings have oil holes or oil troughs which 
catch the oil from the spray and guide it into the bearing surfaces. 

In some small steam engines, in motorcycle engines, and in 
certain types of automobile engines, the crank disk or the flywheel 
revolving inside the crank chamber may be arranged so that it 
dips into the oil, and as the oil adheres to the revolving rim an 
oil spray will be produced. Oil wells or pockets may be cast on 
the inside of the casing, collecting the oil and assisting it through 
various channels, tubes, or troughs, in reaching all parts. 

If the oil level is too low, too little oil spray will be formed ; some 
of the parts will be starved, resulting in inefficient lubrication. 
If the oil level is too high, too much oil spray will be formed, which 
always results in waste of oil, the oil spray escaping from the bear- 
ings or from the air vent usually provided in the crank chamber. 

Excessive oil spray in the case of automobile engines, motor- 
cycles, and other internal-combustion engines is detrimental, 
producing excessive carbonization on the hot pistons. In the 
case of vertical steam engines, excessive oil spray means that 
too much oil passes the pistons and finds its way through the 
engine with the exhaust steam; this means always waste and 
sometimes trouble where it is important that the exhaust steam 
should be as free from oil as possible. 

Circulation Oiling. There are two main systems embodying 
the circulation principle, viz.: 

Gravity-feed circulation. 
Force-feed circulation. 

The gravity-feed circulation system is a central automatic 
oiling system for lubricating a number of bearings and parts, 
e.g., the main bearings, crankpins, crossheads, crosshead guides, 
etc., comprising most of the external moving parts in medium- 
or large-size open-type steam engines, gas engines, Diesel engines, 
steam turbines, groups of large shafting bearings, etc. 


Oil is fed by gravity from a top supply tank through a dis- 
tributing pipe and its branch pipes leading to the various bear- 
ings. Adjusting cocks are fitted in these branch pipes so as to 
regulate the oil feeds, and sight feeds are frequently fitted in the 
oil inlet or outlet pipes to the bearings so that the oil feeds are 
clearly visible. Sometimes, as in the case of steam turbines, the 
sight feeds are fitted in the outlets from the bearings, showing 
the amount of oil that has passed through the bearings. Having 
done its work, the oil drains back from the various parts through 
return oil pipes to a bottom receiving tank. The oil pump driven 
by the engine takes the oil from the receiving tank and delivers it 
either through an oil cooler or direct into the top supply tank. 
If more oil is delivered to the top tank than is required for the 
bearings, the surplus oil passes through an overflow back into 
the bottom receiving tank. 

Drainpipes are fitted to the top tank and bottom tank to 
enable the operator to drain out water, sludge, or impurities 
when required; also the whole or part of the contents of the 
tanks may be withdrawn for treatment in a separation and filtra- 
tion plant. 

It is always difficult to avoid some loss of oil. Oil is lost in 
the form of oil spray, particularly when the speeds are high, 
and is wasted through tiny leaks difficult to avoid and often 
difficult to locate. The loss of oil can be reduced somewhat 
by reducing the amount of oil fed to each bearing, but this is 
doubtful economy, if the lubrication becomes less efficient; 
sufficient oil should be fed so that a good oil film will be main- 
tained, and friction and wear reduced to the minimum. 

A heavy-viscosity oil will cause less loss by leakage or oil 
spray than a low-viscosity oil; but here, again, the bearing fric- 
tion is usually increased, so that from an oil-loss point of view 
very viscous oils should be introduced only if the leakage losses 
are quite abnormal. It pays to provide good save-alls and 
splash guards, not only to save oil but also to save the founda- 
tions. Oil-soaked parts of a foundation are weak and crumbly 
and a constant source of danger to the engine. 

The force-feed circulation system operates on lines exactly 
similar to the gravity-feed circulation system, the difference being 
that the top tank is omitted and the oil passes direct into the 
distributing pipe, which should preferably be fitted with an 


adjustable relief valve, a portion of the oil being by-passed back 
into the bottom tank. The oil is thus delivered under pressure 
as direct as possible to the various bearings and parts requiring 

This system is largely employed for lubricating all sizes of 
enclosed-type steam engines, Diesel engines, vertical kerosene 
engines, gasoline engines, steam turbines, etc. 

Daily Treatment. In the cases of both splash oiling and oil 
circulation it is good practice to remove 2 to 6 gal. of oil every 
day for treatment in a heated separating tank to separate out 
water, sludge, and impurities and afterward to pass the oil 
through a good filter; the purified oil, mixed with a little fresh 
oil, should be returned to the system at the same time that a cor- 
responding quantity of oil is removed from the system for treat- 
ment. When the oil-tank capacity in the system is small, it is 
particularly desirable to recommend this practice. In this way 
the vitality of the oil is kept up to as high a standard as possible, 
and the life of the oil is greatly increased. 

In very large plants, the separation and the filtration apparatus 
are preferably constructed as a part of the circulation system, so 
that either the whole of the oil in circulation or a certain percent- 
age of it is constantly passed through the purifying apparatus. 

Care of Oiling Systems. Whatever oiling systems may be 
employed, it is important that the necessary attention be given 
to institute a regular routine for maintaining the oiling systems 
at their highest efficiency. 

Bearings that are hand oiled should be oiled at sufficiently 
frequent intervals to ensure the presence of an oil film and pre- 
vent excessive heating. The oil containers in siphon oilers, 
bottle oilers, sight-feed drop oilers, and mechanically operated 
lubricators should be filled at correct intervals, and a regular 
system should be employed for putting the oilers into and out of 
service as may be required. Lubricators never should be allowed 
to run empty or to become choked with dirt, and they should be 
cleaned occasionally. 


Reaching the bearing, the oil is conducted to the bearing sur- 
faces through drilled holes ; in order to prevent oil from being 
wasted between the bearing cap and brass, a tube should be 



tightly fitted at this point. The edges of the brasses at the side 
where the oil enters should be chamfered, so as to facilitate the 
entrance of the oil to the bearing surface. This is of paramount 

In bearings employing the ring-, splash-, and circulation-oiling 
systems, where the bearings are 
copiously supplied with oil, oil 
grooves are nearly always 
detrimental; there is usually 
only an oil-distributing groove, 
which runs nearly the whole 
length of the bearing. This 
oil-distributing groove should 

be on the side Of the bearing F IQ . 20. Oil grooving a large crankpin 

where the direction of the bearing. 

revolution of the shaft is downward, and its lower edge should be 

chamfered so as to facilitate the entrance of the oil. 

In bearings that are hand oiled or lubricated by a drop-feed 
system, in which only a moderate supply of oil is introduced into 
the bearings, and where a perfect oil film does not exist, it some- 
times becomes desirable not only to have an oil-distributing 
groove but also to have other suitably cut oil grooves to dis- 
tribute the oil to the bearing surface. 

Under the influence of the bearing pressure the oil is squeezed 
toward the edges of the brass; if the surface speed is high, only 
a small portion will escape, and the loss is replaced at the point 
where the oil enters the bearing. If the surface speed is low, the 
oil received by a certain part of the journal gets time to escape 
and leave the journal surface unlubricated long before that 
particular point has completed a revolution and can receive more 
oil. It is under these conditions that a very viscous oil of good 
body should be used and that oil grooving is desirable. The 
oil grooves should be so cut as to feed oil to several points in the 
bearing and so renew the oil film at these points. Oil grooving 
is frequently much overdone. Cutting large oil grooves removes 
the bearing surface which supports the shaft; it is only in large, 
slow-speed, heavy-duty bearings that oil grooving may become 

Figure 20 illustrates oil grooving in a large crankpin bearing. 
The oil is introduced at the top, and the action of the oil grooves 


is partly to distribute the oil and partly to guide it back toward 
the middle of the bearing, in order to prevent it from escaping 
too freely over the ends of the bearing. 

Oil grooves should always be cut shallow and have rounded 
edges; they should not come too close to the end of the bearing 
brasses; if they are cut close to the ends, oil runs away too freely, 
is wasted, and the bearing will be inclined to heat. 


The frictional heat developed in a bearing spreads into the 
journal and into the bearing itself. Where bearings are not 
water cooled or lubricated by a circulation oiling system, the 
whole of the heat developed must leave the bearing or journal 
by radiation into the atmosphere. Bearings, therefore, assume 
a temperature higher than the surrounding room temperature, 
and the higher the friction the greater will be the difference 
between the temperature of any part of the bearing and the 
room temperature. The difference is termed the frictional rise 
in temperature, or simply the frictional temperature, and forms a 
true guide as to the quality of the oil in service. Any reduction 
in the frictional temperature brought about by * introducing 
another lubricant will mean that this lubricant is better in 
quality or more suitable for the conditions. 

The frictional temperature remains practically constant for 
all room temperatures; i.e., if the bearing temperature is 86F. 
and the room temperature is 70F., the frictional temperature 
is 16F. If the room temperature rises to 74F., it will be found 
that the bearing temperature will rise to 90F; the friction 
developed is practically the same, and the bearing temperature 
must therefore be correspondingly higher, in order to radiate the 
same amount of heat into the atmosphere. 

When bearings operate under conditions of high speed or 
pressure the heat developed may become so great that it cannot 
be radiated from the bearing surfaces sufficiently rapidly. Under 
such conditions it becomes desirable or necessary to introduce a 
circulation oiling system by which the flow of oil going through 
the bearings not only serves to lubricate but also removes a 
large portion of the heat developed, so that this heat, carried 
away with the oil, can be radiated into the atmosphere from the 


oil tanks, oil pipes, etc., or, if necessary, can be removed by an 
oil-cooling arrangement, as in steam turbines. 

Where trouble occurs, it is usually indicated by a tendency of 
the bearings affected to heat up. It will be instructive to analyze 
a number of the causes leading to heated bearings. 

When the barrels of oil have been delivered, it is important 
that they be stored under cover; they should not be left in the 
open, exposed to sun and rain, as, particularly if the barrels are 
stood on end, rain water will find its way through the staves, 
resulting in the glue coating on the inside of the barrel's being 
dissolved and spread throughout the oil. When such oil is used, 
the presence of lining material will cause excessive heating in the 

When opening a barrel, the bung should be loosened by striking 
the staves with a mallet; if an auger is used, fine chips of wood, 
and dirt from the outside of the barrel, may easily find their way 
through the opening into the oil. The oil should therefore always 
be poured through a strainer into the oilcans. If this is not done, 
the small chips of wood and other impurities may get into the 
bearings and cause trouble. 

When the oil is given out from the barrels direct, the overflow 
oil runs on to the floor or into save-alls, which are not always clean, 
and there is the danger that some of this oil, including the dirt 
present, will be given out for lubrication. 

It is good practice to keep the oils in cabinets, preferably pad- 
locked, so that they are not interfered with by unauthorized 
persons; there is then no waste. 

Dirty oilcans are responsible for many hot bearings, and cans 
should therefore be kept scrupulously clean; they should be 
closed at the top or provided with covers, so as to prevent, as 
far as possible, the entrance of dirt. 

An oilcan should never be used for more than one class of oil, 
and in order to prevent mistakes the name of the oil should be 
marked on the can. 

Numerous hot bearings have been caused by the use of 
wrong oil. If, say, a spindle oil is used instead of an engine oil, 
it will cause heating, because it is too light in body to provide 
lubrication. If a very heavy oil is used in place of spindle oil, 
it will cause heating, and the fluid friction will be excessive, 


because it is too heavy to spread over the bearing surfaces, owing 
to the high speed at which the spindles operate. 

In some cases, oils like linseed oil or turpentine have been used 
by mistake; in other cases, the use of badly filtered oil or waste 
oil y instead of fresh oil, has caused great trouble. 
When hand oiling is employed, bearings will be inclined to 
heat if the oiling s are not sufficiently frequent. 

When drop-feed oiling is employed, many hot bearings are 
caused by the lubricators running empty, particularly when the oil 
containers are of small capacity. Sometimes bearings heat 
because the oil congeals in the lubricator or in the feed pipes and 
does not reach the bearings. 

Sometimes parts of the lubricator or the oil-feed pipes from 
the lubricator to the bearings become choked up with deposits of 
various kinds, which may cause a reduction in the oil feed, 
reducing it to such an extent that the bearing heats. 

Fine sawdust in sawmills or woodworking shops, flour dust 
in flour mills, lint in cotton mills, etc., have been responsible for 
such trouble. In one case the sight-feed drop oilers were invaded 
by thousands of tiny little flies, which, after a while, completely 
choked the feed pipes from the lubricators to the bearings. 

Cotton waste, still largely used for cleaning down engines and 
machinery, should not be used for this purpose, as fine fluffy 
matter from the waste gets into the lubricators and oil, causing 
trouble. Mutton or silk cloths are much to be preferred, as they 
are free from fluffy matter and can be readily cleaned. 

Oil may escape between the bearing keep and the bearing 
brass, instead of entering the bearing. With a liberal oil feed, 
the bearing will give no trouble; but when even a small reduction 
in the oil feed is attempted, the bearing will heat, as it is only 
the surplus oil that reaches the bearing itself. 

Very long bearings sometimes give trouble if they have too 
few entrances for the oil. For instance, a bearing more than 10 
in. long and having only one oil inlet by the drop-feed method, 
in the center, will always be inclined to give trouble. 

Some bearings are difficult to lubricate because the pressure is 
upward, instead of downward, which makes it difficult for the 
oil to spread, unless it is introduced at the bottom of the bearing. 

In the case of ring-oiling bearings, water of condensation from a 
very moist atmosphere may enter and accumulate in the bottom 


of the bearing and will lift the oil out of the bearing, until finally 
the oil rings revolve in water, and heating occurs. In ring-oiling 
bearings, deposits formed by the oil itself or by impurities entering 
the bearing may cause the oil rings to stick, so that the oil supply 
fails and the bearing heats. 

Bearings lubricated by the splash oiling system may heat, 
owing to the oil level's being too low to provide adequate oil spray 
or owing to emulsification of the oil by the presence of water of con- 
densation and cylinder oil coming from leaking glands. 

Water, from the engine itself, e.g., condensed steam from 
leaking piston-rod glands, or leaking cooling water, etc., may find 
its way into the bearings and displace the oil; the bearings start 
heating as soon as the oil film is destroyed by the water. 

Where the entrance of water cannot very well be avoided, the 
system of daily treatment of the oil (see under " Turbines") will 
always bring about an improvement. 

In circulation oiling systems bearings may heat because of 
deposit choking the oil-inlet pipes. 

Deposits may be due to unsuitable or improperly manufac- 
tured oil or to the mixing of water and oil or of two different 
oils. If, for example, an oil heavily compounded with blown 
vegetable oils gets into the mineral oil in circulation, a large 
portion of the compound will separate out in the form of a sludge. 

If mineral oil has been a long time in circulation and has 
become very dark in color and considerably weakened, the addi- 
tion of a large quantity of fresh oil will throw down a dark-colored 

Oil-distributing grooves or oil grooves in the bearings may be 
choked up for various reasons already given and thus cause 
trouble, in preventing the proper distribution of the oil. 

Speeding up of the machinery, in order to increase production, 
may cause heating, as obviously higher speed will produce higher 
friction and may demand the selection of a more quick-acting or 
higher quality oil to give good results. 

If the load on an engine is increased, it is not unusual to find 
that some of the bearings are not able to sustain the increased 
strain and, therefore, heat. 

Excessive strains in the bearings may also be produced by the 
settling of foundations, which throws the bearings out of 


Excessive vibration may produce similar results. 

Light load on a steam engine may cause heating of the crank- 
pin bearing, there being an insufficient quantity of steam in the 
cylinder properly to cushion the movement of the heavy piston, 
so that the crankpin is subjected to excessive pressures. 

Eccentric straps may heat on account of bad internal lubrication, 
which increases the resistance in moving the steam or exhaust 

Driving belts and ropes after a time become slack and must be 
shortened. If they are shortened too much, they produce 
excessive pressure on the bearings supporting the pulleys over 
which the belts or ropes run. 

Excessive moisture in the atmosphere causes cotton belts or 
ropes to shrink, whereas leather belting stretches. 

In textile mills where a number of the high-speed spindles are 
operated by cotton tapes and bands, the shrinkage of the cotton 
due to excessive moisture puts excessive pressure on the spindle 
bearings and causes heating. 

Increased temperature will thin the oil, so that it may not be 
able to withstand the bearing pressures; for example, a new addi- 
tion to a boiler plant in close proximity to the powerhouse 
increased the temperature of an engine room so much that all 
bearings heated until an oil heavier in body was introduced. 

Excessive load on an electric motor or the electrical part's being 
out of order will cause high temperature in the rotor; the extra 
heat thus conducted into the bearings may cause the oil film to 
break down, indicated by excessive heating. 

In many classes of rough machinery, it is still frequent prac- 
tice to replace bearings without any attention's being given to 
scraping them together with the shafts; in fact, the bearings are 
allowed to "run themselves in," developing considerable heat and 
necessitating a liberal feed of heavy-bodied oil during the first few 
days. Needless to say, this is a crude and undesirable practice. 

Whenever a bearing has been excessively hot, the bearing 
brasses warp, the cheeks of the brass closing against and nipping 
the shaft; it is necessary to file away and chamfer the edges so 
as to facilitate the entrance of the oil. 

Cracked bearing brasses allow the oil to leak away; the oil film 
is destroyed, and even with a liberal oil feed the bearing will be 
sensitive and inclined to heat. 


Too soft white metal often causes heated bearings, as it yields 
to the pressure and slowly flows out of the bearings, so that the 
bearing surface constantly changes and never assumes a good 
working skin. 

Too hard bearing metal frequently results in heating, because 
the bearing pressures are not uniformly distributed over the 

Rebabbitting of a bearing should be done in one pouring; if 
done in two, the white metal already in the bearing will have 
partly solidified and will not melt properly together with the 
white metal poured in last. The result will be that in operation 
cracks will develop, and the white metal will break loose. This 
also occurs when the white metal has been poured too cold, as 
it does not adhere closely to the shell. 

After a bearing is rebabbitted, the bearing edges should be 
rounded off, and all necessary oil holes and distributing grooves 
properly made. Failure in these respects will cause heating of 
the bearing. 

If appreciable wear takes place, the edges of the oil grooves 
become sharp and act as oil scrapers rather than oil distributors. 
The edges must be kept well rounded, and the oil grooves should 
therefore occasionally be examined, particularly if trouble has 

When worn bearing brasses have been replaced, the bearings 
sometimes heat because the new brasses have not been properly 
fitted or scraped together. 

With crankshafts and the like which have recessed journals 
for the main bearings provided with filleted corners, heating 
may occur if the shaft has insufficient room to float sideways, as 
the shaft will bear hard against the fillet; expansion of the shaft 
may be the cause of this kind of heating; another cause is men- 
tioned on page 167 for electric dynamos. 

If the bearing clearance is too small, through too close adjust- 
ment, heating will occur, as there is insufficient room for the oil 
to produce a satisfactory film, and it becomes difficult for the 
oil to spread. 

If the adjustment of a bearing is too loose, the oil escapes from 
the bearing too freely; and particularly in the case of bearings 
like crankpin bearings, which are subjected to intermittent 
heavy pressures, the oil will not be able to give sufficient cushion- 


ing effect to prevent metallic contact; pounding or knocking of 
the bearing takes place, resulting in heating and wear. 

In starting up after a stoppage, say over Sunday, certain bear- 
ings may be inclined to heat, as the power necessary to drive 
the mill or works is always a good deal higher than normal. 

When engines and machinery have been shut down for a longer 
period, very special attention should be given to the lubricators 
and lubrication of all parts before starting operation again; 
driving belts and ropes are stiff after the long standstill, and it 
must not be expected that the plant can be quickly run up to 
speed without trouble. 

Excessive deflection of a shaft due to various causes will result 
in overheating of the nearest supporting bearings, as the shaft 
will bear more heavily on one side of the bearings, the heat 
developing and spreading from here. 

When bearings of electric motors or generators wear, the slight 
lowering of the rotor due to this wear will cause the magnetic field 
to exert a strong downward pull on the rotor, thus increasing the 
tendency to wear and causing excessive heating. 

Where oils of vegetable or animal character, or at least heavily 
compounded oils, have been used, and where the new oil intro- 
duced is straight mineral or nearly so, the change-over should 
take place gradually, as vegetable and animal oils produce a 
sticky, varnish- or rubber-like coating all over the bearing surfaces 
and in the oil pipes. // the change is made quickly, heating is 
bound to occur or even seizure of the bearing surfaces, as the 
coating is loosened in lumps or flakes, preventing proper oil- 
film formation. It takes time for the bearing surfaces to adapt 
themselves to the new oil. 

When introducing a new oil that is appreciably different in 
character from the oil previously in use, it will nearly always be 
found that some bearings heat up. This may be due to a mineral 
oil's dissolving deposits produced by a compounded oil, which on 
being too quickly loosened cause trouble, acting in the same way 
as grit or dirt. 

The use of a grease containing dirt (which is not visible, as in 
oil) and impure graphite tends to choke oil pipes and oil grooves 
and is often responsible for heated bearings. 

It is not unusual to find that a number of bearings in a mill are 
using an oil far too heavy, because a few bearings, operating under 


bad mechanical conditions, have demanded its use to prevent 
overheating. It would be better economy to use the heavy oil 
on these few bearings only or, better still, to correct the mechan- 
ical conditions so that the proper grade of oil can be used 

Cooling Heated Bearings. When small bearings heat up, they 
are usually easy to cool down, as the total amount of heat present 
in the bearings is not very great ; usually, a liberal supply of the 
oil in use is all that is required; if the bearing is heated to such 
an extent that it has been distorted or the white metal has started 
to flow, it must be dismantled and put in thorough working 

When large bearings heat up, the case is very different, as large 
bearings may absorb and contain a great deal of heat ; and when 
once a large journal starts heating and expanding, there is rela- 
tively so little clearance that the oil film is easily squeezed out, 
and the bearing may seize. The first thing to do when a large 
bearing heats up is therefore to increase the bearing clearance 
by slacking the bearing brasses. 

If the bearing has not seized but is only extremely hot, it is 
usually sufficient to feed it with a liberal supply of steam-cylinder 
oil (which possesses superior lubricating properties under high 
temperature) until the bearing cools, when gradually the normal 
practice of oiling the bearing can be reinstated. 

If the bearing has begun to seize, a little graphite, talc, flowers 
of sulphur, white lead, salt, Sapolio, or like ingredients mixed 
with cylinder oil may be used, as these solid substances help to 
smooth down the parts that have started to cut, thus enabling 
the cylinder oil to form a film. Even more drastic " remedies" 
like brick dust or grindstone dust have been known to cool bear- 
ings, when more greasy ingredients failed to separate the surfaces. 

Castor oil is often employed for cooling bearings but should be 
avoided where a circulation system is employed, because it 
mixes with the engine oil and afterward develops deposits. 
Once a bearing has become accustomed to the use of castor oil, 
it is not always a simple matter to change back to the original 

The practice of using water for cooling the bearings from the 
outside is very undesirable, as the result of the sudden cooling 
is nearly always distortion of the bearing brasses, so that they 


have to be filed and scraped before satisfactory operation can 
again be expected. 



The object of bearing lubrication is (1) to form a lubricating film 
between the rubbing surfaces and thus replace the metallic friction 
with fluid friction, as far as possible; (2) to reduce the fluid friction 
in the oil film itself to the lowest safe point, considering the 
operating conditions. 

No Lubrication. If a journal revolves in its bearing without 
lubrication, metallic contact will cause abrasion of the metal, and 
the bearing will not operate very long before the frictional heat 
developed will be so great that the bearing surfaces will be 

Oilless bearings are an exception; they are made of a metal 
alloy or compressed wood, mixed with graphite, talc, or other 
solid lubricant; or the graphite is firmly placed in the bearing in 
the form of spiral grooves or strips; or, again, the whole bearing 
may be compressed talc, soapstone, or graphite. Such bearings 
will often run without lubrication and without seizure, but the 
friction is very high, as is also the bearing temperature. 

Boundary Lubrication. By introducing a lubricating medium 
between the rubbing surfaces, the lubricant will adhere to the 
journal as well as to the bearing, thus replacing part of the 
metallic friction with fluid friction; there will be less abrasion, 
therefore less wear, friction, and heat. 

The vast majority of bearings are boundary lubricated; i.e., the 
rubbing surfaces are never kept completely apart, so that more 
or less wear does occur, and the loss in friction is not so low as it 
might be. 

As all fixed oils are more oily than mineral oils, an admixture 
of a small percentage to the mineral oil will increase its oiliness 
and assist in separating the rubbing surfaces more completely. 

If it were not for the high price of fixed oils and their tendency 
to gum (particularly the vegetable oils), they ought to be much 
more widely used than they are at present. It is particularly 
for heavy pressures and slow speed that great oiliness is so very 
desirable, necessitating the use of fixed oils. It is a well-known 
fact that castor oil and rape oils are extremely useful for very 
severe conditions of this kind. 


Compounded oils also have the property of combining and 
emulsifying with water, so that their use is desirable where water 
gains access to the bearings. Water will displace a straight 
mineral oil and cause trouble but will combine with a com- 
pounded oil and form an emulsion or lather, which, particularly 
in the case of marine steam engines, is very desirable. If a 
bearing under such conditions heats up, the lather escaping 
from the bearing will lose its milky appearance and become 
semitransparent, this being an indication of excessive bearing 

Compare remarks under " Textile Machinery/' " Marine Steam 
Engines," " Locomotives," " Stainless Oils," etc. 

Oil-film Lubrication. By introducing a sufficient quantity of 
oil it is possible to form between the rubbing surfaces a complete 
oil film, which means that there will be no wear and that the 
friction developed is reduced entirely to the fluid friction within 
the oil itself. Given the necessary surface speed, a suitable 
bearing pressure, and the required flow of oil, as will often be 
the case with circulation-oiling, ring-oiling, and bath-oiling sys- 
tems, the friction is entirely fluid friction determined by the 
viscosity of the oil, the surface speed, and the area of the rubbing 
surfaces; oiliness is of no importance (except when starting and 
stopping) ; the viscosity alone is what maintains the oil film. The 
higher the viscosity the more easily will the film be formed at 
low speeds; but at high speeds, high-viscosity oils may give 
trouble, and low-viscosity oils should always be preferred. 


In order to obtain efficient lubrication, oils must be selected to 
suit the operating conditions and the oiling system employed. 

Operating Conditions. The oil must be selected to suit the 
conditions of size, speed, pressure, temperature, and mechanical 

Speaking generally, oils light in body should be employed for 
such conditions as small bearings, high surface speed, low bearing 
pressure, low room temperature, and good mechanical conditions. 

Speaking generally, oils heavy in body should be employed for 
large bearings, low surface speed, high bearing pressure, high 
room temperature, and bad or indifferent mechanical conditions. 


Oiling Systems. The oil must also be selected to suit the 
oiling system employed. 

Hand Oiling. Hand-oiled bearings are rarely well lubricated; 
they are usually only boundary lubricated and demand the use of 
heavier bodied oils than would be required with a more efficient 
oiling system. This system wastes both oil and power. Unless 
the waste of oil is very abnormal, compounded oils should be 
preferred for hand oiling, as such oils have greater oiliness than 
straight mineral oils, therefore last longer and give less 

Drop-feed Oiling. In drop-feed oiled bearings, less oil is wasted 
than in hand-oiled bearings, and, owing to the more regular oil 
feed, the oil film in the bearings is kept more uniform and more 
complete; the lubrication is therefore more efficient; i.e., there is 
less friction and less wear. Under high-pressure conditions, com- 
pounded oils should preferably be used; for low or moderate 
pressures straight mineral oils will render good service. 

Ring Oiling. By the ring-oiling system the bearing surfaces 
are constantly flooded with oil, so that the lubrication becomes 
as efficient as possible with the particular grade of oil in use. 
Straight mineral oils should be used, as compounded oils will 
cause gumminess on the oil rings and in the bearings. 

Splash Oiling. The oil should be light in body so as to splash 
easily to all parts yet sufficiently heavy to produce satisfactory 

Circulation Oiling. As the oil is forced in large quantities to 
the bearings, it is given every assistance to produce complete 
and perfect lubrication, and the heat is so rapidly removed that it 
becomes possible to operate engines employing this system at the 
highest speeds and yet maintain a great margin of safety in opera- 
tion. The oil must, however, be of such a character as to main- 
tain its nature, notwithstanding that it circulates continuously 
and is exposed to the oxidizing influence of air and impurities, 
the emulsifying influence of water, etc. Also, it must be of 
such a nature as to separate quickly from water and impurities, 
so that sludge or deposit developed may be easily removed from 
the oil in circulation. As to the nature of such oils circula- 
tion oils see remarks under " Turbine Lubrication" (page 243). 

The best oils used in splash oiling, ring oiling, or oil-bath 
systems possess similar characteristics. 


Where hand oiling or drop-feed oiling systems are employed, 
the oil, after passing through the bearings once, is frequently run 
to waste and not used over again, in which case the slight altera- 
tion that takes place in the oil passing through the bearing is of 
no importance, and compounded oils can be used without trouble. 

When the oil, after passing through the bearings, is collected 
and filtered for the purpose of using it over again, either on the 
same bearings or for less important work, mineral oil may be 
preferred, particularly if it be used over and over again a great 
number of times on important bearings and with only slight loss. 

Selection of Oil. It will now be understood that when select- 
ing oils for bearings operating at high speed, with low bearing 
pressures, and employing a good lubricating system, the chief 
object should be to reduce the fluid frictional losses, as here the 
question of wear is less apt to become an important factor. 

For high-speed spindles in textile mills, high-speed shafting, 
and machinery of many types, oils of the correct light body and 
quality should therefore be selected, and the result will be an 
appreciable reduction in power. 

In bearings operating at slow speed, with heavy bearing pres- 
sures and using a less efficient oiling system, the danger of wear is 
great, and the chief object of lubrication here becomes minimiza- 
tion of wear, rather than the reduction of fluid friction. 

For such bearings as are employed in large open-type steam 
engines, heavy pumping plants, and heavy machinery bearings, 
oils of the correct heavy body and quality should be selected. 

There are many plants in which it is declared that there is no 
trouble. Whether this be so or not, there is a long distance from 
this no-trouble standpoint to perfection in operation; it is only 
by analyzing the actual conditions, carefully grouping various 
portions of the machinery and using specially selected oils for 
each group to give maximum lubrication service, that perfect 
results can be obtained and maintained. 

There are many types of modern machinery, such as steam 
turbiiues, high-speed steam engines, internal-combustion engines 
of all kinds, and other high-speed machinery, where the condi- 
tions demand the use of the highest quality oil obtainable, almost 
regardless of its cost, and where smooth and safe operation and 
low frictional losses count many times more than the cost of the 
oil itself. On the other hand, where the class of machinery 


in use is rough or in bad repair, where wasteful and inefficient 
oiling systems are employed, and particularly where the care and 
attention given to the plant are indifferent or bad, it is not always 
possible to justify the use of the best quality lubricating oils. 
So much oil may be wasted to no useful purpose that the cost of 
the oil thus literally thrown away will more than outweigh the 
value of the better lubrication that might be brought about 
by the use of better oils. 


To satisfy the bearing requirements of the great variety of 
engines and machinery in existence, a great number of bearing 
oils are needed. Many of these oils will be mentioned under the 
class of machinery for which they are recommended, e.g.: 

Circulation oils For steam turbines, enclosed-type steam 

engines, etc. 
Marine-engine oils For marine steam engines and other severe 

Loco engine and car oils .... For locomotives, tenders, and cars 

Spindle and loom oils For textile machinery 

Black oils For mine cars and rough machinery 

Steam-cylinder oils Used for bearing lubrication of enclosed-type, 

splash-oiled steam engines 

In all these cases there are service conditions that call for 
some special property in the oil and therefore justify grouping 
such bearing oils in the way indicated above. 

With bearing oils the author proposes to refer to oils intended 
to be used and recommended for all types of machinery, where the 
service conditions do not present any specially difficul features. 

In other words, bearing oils are oils whose prime duty it is to 
lubricate and which are not required to withstand oxidation or 
emulsification (as circulation oils) or to lubricate heavy bearings 
in the presence of water (as locomotive- and marine-engine oils) 
or to possess stainless properties (as loom oils), etc. 

Bearing oils are oils ranging in color from light to deepest 
red; they must be refined but need not be specially well refined; 
in fact, excessive acid treatment or earth filtration removes many 
active unsaturated hydrocarbons, some of which are quite as 
good as if not better lubricants than the saturated hydrocarbons. 



A certain degree of refining is, of course, needed to remove a 
sufficient amount of the most unsaturated elements which, if 
left in the oil, would cause excessive gumming in the bearings. 

The oiliness of distilled mineral lubricating oils can be improved 
by admixture of a small percentage, say from 5 to 10 per cent, of 
fixed oil or a certain percentage of filtered cylinder stock. To 
make this point clearer, the author has found that when using 
oils compounded in this manner, (i.e., admixture of fixed oil or 
filtered cylinder stock), lower viscosity oils can be selected to 
render certain service than if a distilled mineral lubricating oil 
were to be employed; the result is, therefore, lower friction and 

Such savings in power accomplished by using lower viscosity 
compounded oils are mentioned on page 330 for textile machinery. 
In the same way, power savings can be obtained by replacing a 
distilled mineral oil of a certain viscosity by a lower viscosity oil, 
which is made by mixing a spindle oil (or a blend of a spindle oil 
and a medium red oil) with a certain amount of filtered cylinder 

Without going more deeply into this matter the author gives 
below approximate viscosities (see table, page 57) for six bearing 
oils, which will be found to cover a wide range of service. 

For Bearings 


oil num- 


absolute vis- 
cosity, in 

Recommended for 



at 50C. 




Very light duty and high speed 




Light or medium duty and medium 

or high speed 




Medium duty and medium or high 





Medium or heavy duty and medium 

or high speed 




Heavy duty and slow or medium 





Heavy duty and slow speed 



The author hesitates to give the foregoing service recommenda- 
tions, which of necessity are very crude, but under the various 
sections on engines and machinery following this chapter he has 
endeavored to convey his ideas and experience in a more definite 


The various methods by which semisolid lubricants are applied 
may be classified as follows: 

Contact feed. 
Stauffer cups. 
Compression cups. 
Mechanical feed. 
Grease bath. 

Contact Feed. By this method the grease is placed direct on 
the journal, e.g., in the dryer bearings of paper machines and the 

roll-neck bearings in steel mills. 
The grease adheres to the 
journal, melts away or softens, 
and gradually wears away! 
Hard greases are generally 
employed. With soft greases 
the consumption is usually 
great, particularly when the 
bearing is worn, as in that case 
the grease adhering to the jour- 
nal is pulled into the large 
clearing space between the 
journal and the keep and is 


21. Contact-grease-feed arrange- 

quickly consumed. 

When soft grease is applied direct to the shafting it must be 
protected by a layer of yarn-fiber grease, e.g., in shafting bearings 
for weaving sheds and in bearings used in connection with the 
rollers that support rotary kilns in cement works. Such bearings 
have a large cavity at the top (Fig. 21); the yarn grease is placed 
all around the walls of this cavity, and sometimes also there is 
a bottom layer touching the journal. In the pocket thus formed 
is placed ordinary cup grease or fiber grease, the grade being 
selected in accordance with the temperature conditions; exposed 



to the heat, the grease in the pocket melts slowly through the 

yarn grease and lubricates the journal. 

Figure 22 shows a gravity-feed grease cup designed for the use 

of low-melting-point greases or oils, which are slightly soap 

thickened, so as to be nonfluid at ordinary room temperatures. 
The needle, in touching the shafting, gets warm, melts a little 

of the grease and acts very much like 
the needle in glass bottle oilers (Fig. 
17, page 115). 

Stauffer Cups. Figure 23 shows an 
ordinary plain cup; the bottom is 
preferably sloping to facilitate the 
grease's being forced out of the cup. 
The cover is given an occasional turn, 
and a quantity of grease is forced 
into the bearing; it is gradually con- 
sumed until the cover is given another 

FIG. 22. Gravity-grease-feed 

FIGS. 23-24. Stauffer cups. 

turn. To prevent thin grease from leaking out, the thread must 
be a good fit, or a leather packing must be introduced as shown in 
Fig. 24. This drawing also shows a catch pawl which prevents 
the cover from slacking back. 

Compression Cups* Compression cups may be operated either 
by a spring or by compressed air. Figure 25 shows a typical 
spring compression cup. The spring (1) pushes against the piston 
(2). The feed can be adjusted by the screw (3). Only greases 
of No. 1 and No. 2 consistency can be used in this type of cup. 

For. harder greases of No. 3 or No. 4 consistency, a grease cup 
must be used like Phillips crankpin grease cup shown in Fig. 26. 
By turning the milled collar (1), grease is forced up into the small 
cylinder (2), raising the piston against the force of a strong spring, 
which subsequently feeds the grease until the indicator knob (3) 
shows that another feed must be given. 



Figure 27 illustrates the Menno compressed-air cup in which 
compressed air is employed for forcing out the lubricant. The 
lubricant is filled into the bottom portion (1) of the cup; this 
part is threaded to receive the upper portion (2), which on being 
screwed into the lower portion causes a certain air pressure to be 
formed above the grease. The object of the thin metal disk (3), 
which is guided vertically, is merely to rest on top of the grease 
and keep it level. The fixed disk (4) forms the top of the air- 

FIG. 25. Spring com- FIG. 26. Phillips crank- FIG. 27. Menno grease 
pression cup. pin grease cup. cup. 

compression chamber. After giving the upper portion (2) a 
certain number of turns, it is locked to the bottom portion by 
means of a lock nut (5) and the air pressure will maintain a fairly 
regular feed. If the journal gets warm, the heat is conducted 
up into the cup through a thin funnel. The effect of this rise in 
temperature is to soften the grease, increase the air pressure, 
and give an increased feed of lubricant. 

In grease cups for lubricating loose pulleys the centrifugal 
force acting on a piston may be made use of to force thin grease, 
or nonfluid oil, to the bearing. 

Mechanical Feed. In very large colliery winding engines or 
steelworks rolling-mill engines, hard greases, usually white, may 
be forced into the crankpins by means of mechanically operated 
lubricators, as indicated in the sketch (Fig. 28). 

The arrangement is very similar to the banjo oiler. A cam 
on the feed pipe (3) operates a ratchet lever (4). The motion of 



the ratchet wheel (5) and worm gear (6) actuates a piston (7) 
which forces the grease below the plunger through the feed pipe 
(9) into the crankpin. Another method is to place the lubricator 
complete on the crankpin itself. The lever (4) is then weighted 
at its lower end and swings to the right and left between two 
adjustable stops, owing to the motion of the crankpin. The 
lever in this way oscillates sufficiently to operate the ratchet, and 
the feed may be adjusted within certain limits, say one, two, or 
three teeth per revolution. 

The advantage of a mechanical feed as against compression 
cups is that the lubricant, whether soft or hard, is delivered 


t " 

1 Connecting Rod 

2 Crank Pin Bearing 
B Driving Cam 

1 Hatchet Lever 

5 Ratchet 

6 Worm Wheel 

7 Piston 

Grease Feeding Pipe 

r 3 

FIG. 28. Mechanical grease lubricator. 

absolutely uniformly, notwithstanding changes in temperature 
which either harden or soften the grease and the result of which 
with ordinary grease cups is that a uniform feed cannot be 

Grease Bath. A bath of grease may be employed in connec- 
tion with ball and roller bearings, gearboxes of automobiles, 
gear chambers in pneumatic tools, etc. 

The reasons why grease is employed are outlined under these 
several headings and are mainly to keep dust or grit out of bear- 
ings or to prevent excessive leakage of lubricant. 

Greases of Nos. 1 and 2 consistencies are used for grease baths ; 
harder greases create undue friction, are inclined to cake exposed 
to heat, and do not distribute themselves with sufficient ease. 



The conditions for which semisolid lubricants are advantage- 
ously employed will be indicated in the following, fuller infor- 
mation being given under the various sections of machinery, 
etc., referred to. 

In dusty and dirty surroundings, e.g., cement mills, bakeries, 
colliery screening plants, etc., grease keeps the bearings clean; 
it entirely fills the bearing cavities and the clearance space and 
forms a fillet round the bearing ends, which prevents the entrance 
of dust and dirt. This is particularly important for ball and 
roller bearings. 

When oil is used in weaving sheds, it is necessary to fix save-alls 
below the bearings. For this reason, grease is sometimes pre- 
ferred to oil, because there is less likelihood of the spent lubri- 
cant's dropping from the bearings on to the looms and soiling the 

When bearings are in inaccessible places and cannot be lubri- 
cated with oil by ordinary means, grease cups can be fitted, and 
the grease forced through tubes into the bearings from any angle. 

Greases of high melting point fiber greases and others are 
required occasionally where the bearing temperatures or room 
temperatures are unusually high, such as the hot necks on dryers 
in paper mills, hot journals supporting the rotary kilns in cement 
works, and hot-roll necks in tin-plate mills and steel mills. 

Grease should be used only where there are special reasons 
against the use of oil. Wherever grease is used under conditions 
that are quite suitable for oil lubrication, the introduction of the 
correct grade of oil will always result in an appreciable saving in 
power. Grease lubrication means a heavy frictional resistance in 
the bearings, as obviously the grease does not begin to lubricate 
until the frictional temperature has increased to such an extent 
that the grease melts or becomes sufficiently soft to be " abraded" 
by the revolving journal. 

The suitability of a grease depends on four things: 

1. The purity of the grease and the absence of filling matter. 

2. The consistency of the grease (to suit the method of appli- 

3. The quality of the oil and other ingredients in the grease. 

4. The melting point of >the grease (to suit the temperature 


Purity is very important in greases that are used under con- 
ditions of high pressure or speed. Such greases should prefer- 
ably be strained hot as mentioned (page 25). 

Filling matter is nonlubricating; it lowers the manufacturing 
cost of grease but usually detracts from its lubricating value. 
For rough mechanical conditions or for very high bearing pres- 
sures and slow speed, filling matter like graphite, talc, or mica 
may, however, prove advantageous, helping to fill up unevenness 
in the rubbing surfaces and preventing seizure. Filling matter 
containing gritty or hard impurities will cause wear but may 
prove beneficial as a temporary remedy in the case of hot bearings. 

Consistency. The consistency of grease is largely governed 
by the feeding appliances. If grease is applied through com- 
pression cups or gravity-feed cups, it must be soft, either No. 4 
or No. 2 consistency, also when used for high-speed bearings and 
ball and roller bearings. 

For Stauffer cups, No. 2, 3, or 4 consistency can be employed. 

A grease of No. 4 or 5 consistency may be selected for contact- 
feed application in connection with slow-speed open bearings, 
the grease resting directly on the rotating shaft. 

Quality. The quality of grease depends largely on that of the 
lubricating oil used in manufacture. 

For medium- and high-speed work, with no excessive bearing 
pressures, a grease should be chosen that contains a light-bodied 
lubricating oil. 

For medium- and slow-speed work with fairly heavy bearing 
pressures, the grease should preferably contain a more viscous 
oil; and for extreme conditions of pressure and slow speed, fatty 
oils or fats must form part of the grease, as great oiliness is 
required. A percentage of solid lubricants may also be of advan- 
tage, as mentioned under " Filling Matter." 

Changing Grease. When a change is made from white and 
other greases that contain much tallow or other fat or fatty 
oils to a mineral grease, as cup grease, the process must be 
gradual to avoid heating, this being just as necessary as when 
changing from, say, castor oil to a mineral oil. 


In order to determine the proper range of service for the various 
solid lubricants in the field of lubrication, the subject will be 
divided into the following sections: 


Theory of the Action of Solid Lubricants. 

Methods of Application and Use. 

Observations on Results Obtained by the Use of Solid Lubricants. 

In order to avoid having to refer repeatedly to the use of solid 
lubricants throughout the book, the author has at this place dealt 
with the entire field of service for solid lubricants in such a man- 
ner that further references to this subject may perhaps be con- 
sidered unnecessary. 


It is generally agreed that the friction created in engines or 
machinery of all kinds is composed chiefly of what may be termed 
solid friction or fluid friction or a combination of both, the latter 
condition representing the state of affairs in the great majority 
of cases. 

In the following, the influence of solid lubricants on each of 
these various kinds of friction will be dealt with separately. 
Reference will also be made to the use of solid lubricants for 
treatment of hot bearings. 

Solid Lubricants and Solid Friction. When a solid lubricant is 
introduced between otherwise unlubricated surfaces, the more or 
less finely divided particles of the lubricant associate themselves 
with one or other of the rubbing surfaces, filling in the pores and 
depressions and acting, as far as possible, as a smoothing and 
polishing agent, covering the original surfaces with a thin, smooth 
layer of the solid lubricant. As a result, the coefficient of fric- 
tion is reduced; the solid friction between the more or less rough 
original rubbing surfaces is replaced by the lesser solid friction 
between the smooth surfaces formed by the solid lubricant. 

When abrasion takes place, it occurs not so much between the 
original surfaces (which possess great cohesion) as between the 
particles of the solid lubricant, which have but little cohesion. 
Artificial amorphous graphite, for example, has practically no 
cohesion. If solid lubricants are employed, cutting and abrasion 
of the bearing surfaces are therefore much less likely to 

There are a variety of conditions for which dry solid lubricants 
have proved advantageous, e.g., in bearings or in such parts 
of machinery as are apt to be neglected from a lubricating point 


of view, and that operate at low pressures and low speeds. When 
such surfaces are well coated with graphite, for example, and 
particularly if they are rubbed down to a dense glazed finish, they 
will work upon each other for a long time with comparative 
freedom and without danger of cutting or wear's taking place. 

Solid Lubricants and Fluid Friction. The application of 
solid lubricants to bearings in which a perfect oil film is estab- 
lished would at first sight appear to be of no value; the journal 
floats on a film of oil, and the presence of small particles of a solid 
lubricant does not increase the viscosity to any appreciable 
extent. The friction under running conditions is therefore nof 
increased unless the solid lubricant is present in such an amount 
that the particles "crowd" the oil film at the " point of nearest 
approach" between journal and bearing and start to act as an 
abrasive powder. 

It has repeatedly been noticed in experimental work that 
immediately after a temporary application of solid lubricant in 
powder form, the friction is much increased but is reduced after- 
ward, when the particles have had time to attach themselves to 
the rubbing surfaces and form a smooth coating. The virtue in 
the employment of a solid lubricant lies entirely in the effect that 
it produces on the rubbing surfaces themselves. With perfectly 
lubricated bearings the chief advantage of using a solid lubricant 
is apparently the effect on the friction at the moment of starting, 
which results in a reduction in the static coefficient of friction. 

Static and Kinetic Friction. The effect of the use of a suitable 
solid lubricant or a solid colloidal lubricant is, as we have seen, to 
reduce the tendency to abrasion and to produce a smpothness of 
the surfaces. As the solid lubricant cannot be displaced by pres- 
sure, the static coefficient of friction is reduced as compared with 
the result obtained when oil alone is used, assuming that the 
solid lubricant is of such a nature and used in such a manner that 
it has actually increased the smoothness of the rubbing surfaces. 

Solid Lubricants and Boundary Lubrication. Under these 
conditions there appear to be great possibilities for the use of 
solid lubricants. Their object will be: 

1. To reduce the solid friction. 

2. To produce a smoothness of the rubbing surfaces, which will assist 
in distributing the load evenly over all parts of the bearing and thus enable 
a lower viscosity lubricant to be used and the fluid friction to be reduced. 


3. To reduce the wear of the original surfaces and the risk of abrasion 
or cutting of the surfaces which ordinarily leads to the production of hot 

4. To reduce the consumption of lubricant. 

To obtain these advantages, the solid lubricants must be of 
suitable nature, purity, fineness, and hardness and must be used 
in the right amount. 

Nature. A good solid lubricant must possess ability to adhere 
to metallic surfaces, and it must be capable of producing a smooth 
surface. Graphite possesses both of these properties to a marked 
degree. When rubbed between metallic or nonmetallic sur- 
faces, graphite whether of the flake or of the amorphous variety 
produces a coating that is smooth and unctuous. Talc and mica 
do not adhere to surfaces so well as graphite does, nor do they 
produce so smooth a surface. 

The quality of unctuousness in the surface produced is undoubt- 
edly important; it is not possessed by materials such as flowers 
of sulphur or white lead, which act more as abrasives than as 

Purity. A high degree of purity of the solid lubricant is neces- 
sary in connection with lubrication of all high-class machinery; 
whereas for rough bearings operating under extreme conditions 
and on the verge of seizure, a small amount of impurities may not 
be detrimental. 

Fineness. In the case of well-finished rubbing surfaces, very 
finely divided graphite must obviously be used, and the coating 
is easier to accomplish than with rough surfaces. Under these 
conditions, makers of amorphous graphite claim that a flake 
graphite when used in excess is apt to build up too thick a surface 
and reduce the working clearance to a dangerous extent, whereas 
with amorphous graphite, excessive use can have no ill effects; 
the soft, crumbly, amorphous grains are easily crushed; in fact, a 
surface of fine amorphous graphite under pressure moves within 
itself like a film of oil, and the particles are noncoalescing and offer 
little resistance to movement. 

With highly finished and polished surfaces operating with 
small clearances it would seem undesirable to use powdered 
lubricants, however finely they may be pulverized. Colloidal 
lubricants appear to be the only solid lubricants likely to give 
satisfaction under such conditions. 





Pure graphite 1.0 

Best quality of talc 1.0 

Lower qualities of talc or soapstone 2.5to4.0 

Micas 2.0 to 3.0 

The admixture of a hard solid lubricant, like hard talc or mica, 
to a grease, particularly if an excessive amount is added, may 
cause a great deal of continuous but uniform wear much more 
than would be caused by the grease used by itself yet no cutting 
or excessive heating of the bearing may occur. 

Amount. Makers of flake graphite recommend the admixture 
of 3 to 4 per cent of fine flake graphite with oil; if too much 
graphite is used, the friction is increased, because more graphite is 
introduced into the bearing than is required to keep the rubbing 
surfaces properly graphited. The surplus graphite present 
between the rubbing surfaces creates extra friction and heating. 

If appreciably less graphite is added than 3 per cent, a point 
will be reached when the graphite coating will no longer be fully 
maintained, and the full benefits from the use of graphite will not 
be obtained. 

Makers of graphite greases recommend a percentage of graphite 
ranging from 3 to 10 per cent. More graphite is required with 
grease than with oil, because grease is usually employed for 
rougher conditions than oil, and more graphite is needed to build 
up the surfaces and maintain them in a smooth condition. 

The effect of adding a solid lubricant to a lubricating grease is 
that in time the solid lubricant will attach itself to the rubbing 
surfaces and, by smoothing and polishing them, will make it 
easier for the lubricant to do its duty. As a result, a softer 
grease or a grease containing a lower viscosity oil can be employed 
than when no solid lubricant is added to the grease. 

Makers of colloidal graphite find that a very small percentage 
of graphite is ordinarily required in the diluted colloidal lubri- 
cant. Acheson recommends a graphite content of 0.35 per cent 
for most purposes. That this small amount has been found 
sufficient is probably explained by the fact that colloidal lubri- 
cants are used chiefly on high-class machinery with reasonably 
well-finished bearing surfaces. 


Hot Bearings. Hot bearings may be caused by excessive 
stresses or vibrations, by the accidental entrance of gritty 
impurities, by a shortage of lubricant, etc. Whatever the cause 
may be, the oil film becomes entirely displaced from a small 
portion of the bearing surface; a "dry" spot is formed; the sur- 
faces enter into intimate metallic contact; the local temperature 
rises rapidly; the bearing seizes; and, if it is lined with white 
metal, the latter may melt and flow out. Under such conditions, 
when a bearing gives warning by heating, the usual procedure is 
to resort to the use of a fixed oil, like castor or rape oil, or to a vis- 
cous mineral oil, like steam-cylinder oil; the effect of using such 
oils is to produce a better film, which separates the metallic 
surfaces and reduces the temperature. 

When the surfaces have begun seriously to abrade one another, 
oils may prove of no avail, and solid lubricants must be used, 
such as graphite. The graphite particles by coating and impreg- 
nating the surfaces make it difficult for the metallic surfaces to 
seize; and if slight abrasion takes place in certain places, the 
graphite may often succeed in repairing the damage and make it 
possible for the normal lubricant again to take care of the 

Flowers of sulphur and white lead are often used to cure hot 
bearings; they act not so much as lubricants but rather as mild 
abrasives; they grind away the rough spots and produce a 
smooth surface. 

Much more drastic remedies, such as salt, brick dust, and 
grindstone dust, have been successfully employed in very serious 
cases of large hot bearings; their function is to grind away 
quickly the rough parts that have begun to seize. They may 
be applied mixed with thick steam-cylinder oils or castor oil, in 
order to produce a thick film. The oil should be applied liberally 
in order to clean away the gritty powder after it has done its duty. 

In bearings that are inclined to run hot, it is good practice 
occasionally to apply a small amount of graphite to produce a 
graphitized surface or to mix colloidal graphite with the normal 
lubricant, so as continuously to make up the wear on the graphite 
coating. In overloaded worm gears, for example, which are 
continuously inclined to seize, it is good practice to mix a small 
amount of flowers of sulphur or fine graphite with the oil; they 
serve to prevent seizure, and the wear becomes more uniform. 



Solid lubricants may be applied in three different ways: 

1. Dry application. 

2. Mixture of solid and semisolid lubricants. 

3. Mixture of solid and liquid lubricants. 

Dry Application. Solid lubricants are applied dry in cases 
where for special reasons it is inadvisable or impossible to use 
an ordinary liquid or semisolid lubricant. The finely powdered 
solid lubricant is put into a linen bag, and the bag is pounced or 
struck against the parts requiring lubrication; or a syringe like 
that used for applying insect powder may be employed to inject 
a'cloud of lubricating powder into the bearings. 

The following examples are illustrative: 

Lacemaking Machines. On certain reciprocating parts pow- 
dered graphite is used in place of oil, to avoid staining the fabric. 

Bottle-making Machines; Galvanizing Machines. Certain parts 
are exposed to extremely high temperatures; oil would burn 
away and leave a carbonaceous residue which would cause the 
parts to stick. 

Chocolate and Candy Machinery. To avoid oil's dropping into 
the chocolate or candy, all bearings may be lubricated entirely by 
dry graphite powder. The pressures and speeds are low, so that 
the friction developed is not too great for comfortable running. 

Oilless Bearings. Oilless bearings are referred to on page 130. 

For the lubrication of rubbing surfaces made of wood, graphite 
is very suitable; it is not absorbed, as in the case of oil. The 
graphite may also be applied mixed with grease, for the sake of 
convenience of handling. 

Steam Cylinders and Valves. Dry graphite in the form of small 
cylindrical sticks has been used in conjunction with oil for lubri- 
cating locomotive valves and cylinders, the oil being supplied by 
a separate lubricator. The graphite sticks are placed in a vertical 
tube and rest upon an abrasive wheel, which obtains a rotative 
or oscillating motion from some reciprocating part of the engine, 
e.g., the valve rod. In this way, the abrasive wheel continuously 
abrades the bottom graphite stick, and the graphite powder drops 
down a vertical passage direct into the engine. 

Mixture of Solid and Semisolid Lubricants. The use of a 
solid lubricant in powder form is resorted to only in special 


circumstances. When there is no particular objection to the use 
of a fluid or semisolid lubricant and it is desired to use a solid 
one, it is obviously desirable to mix the two together. Semisolid 
lubricants are eminently suitable as carriers for solid lubricants 
because, being nonfluid, they prevent separation of the graphite, 
and, as they are themselves gradually consumed, they automati- 
cally supply the solid lubricant to the parts that they lubricate. 

The admixture of solid lubricant usually ranges from 3 up to 
10 per cent, rarely exceeding the latter amount. 

Speaking generally, semisolid lubricants are always improved 
by the admixture of a small amount of finely pulverized pure 
flake or amorphous graphite. Exceptions are bearings with 
highly polished surfaces and small clearances and high-class 
ball and roller bearings, for which colloidal solid lubricants are 
the only solid lubricants that can be considered. 

Mixture of Solid and Liquid Lubricants. Ordinary solid 
lubricants cannot normally be applied mixed with liquid lubri- 
cants, because, however finely the solids may be pulverized, 
their high specific gravity causes them to settle out in the lubri- 
cators, oil pipes, etc. The finer the particles and the more 
viscous the oil the more slowly does separation take place, so that 
slight agitation may be sufficient to prevent separation. Mix- 
tures of very finely pulverized solid lubricants and viscous oils, 
such as gear oil for automobile gearboxes, may be kept mixed 
by the stirring motion set up by the gears. 

Certain mechanically operated graphite-oil lubricators for 
steam engines are fitted with stirrers in the lubricator container 
as well as in the oil pipe leading from the lubricator to the 
engine, to assist in preventing the graphite and oil from 

This problem of preventing separation of the solid lubricant is 
one that is causing many difficulties and cannot be said to have 
been satisfactorily solved, on account of the mechanical compli- 
cations involved. 

Chapman and Knowles have patented a mixture of finely 
pulverized graphite and glycerin for lubricating steam-engine 
cylinders. Before being mixed with the glycerin, the graphite 
is impregnated with a sufficient amount of petroleum or other 
hydrocarbon insoluble in glycerin, to reduce the specific gravity 
of the mixture to that of glycerin. As a result, the "graphite- 


petroleum" specks will remain in suspension in the glycerin, 
and the mixture can be pumped by a mechanical lubricator and 
supplied to the steam engine in the ordinary way. 

Solid lubricants can, of course, be mixed with oil and, in the 
form of a more or less liquid paste, may be applied by hand to 
the bearings or parts requiring lubrication. This method is the 
one employed when " curing" hot bearings. 

It would appear that the only really satisfactory way in which 
a solid lubricant can be automatically applied mixed with a 
liquid lubricant is to bring the solid lubricant into such a finely 
divided state that the particles become of a size approximating 
that of submicrons. This state of fineness cannot be obtained 
by mechanical means alone but has been attained by certain 
processes, such as Acheson's process already referred to. Col- 
loidal solid lubricants, when diluted with pure oil (oildag, oleosol) 
or pure water (aquadag, hydrosol) do not separate out to any 
extent; they can be diluted indefinitely and can therefore be 
applied to any engine or machine, mixed with the diluent which 
serves as a carrier. 

Archbutt 1 has made some siphoning tests with oildag and has 
proved that deflocculated graphite will pass over with lubricating 
oil through worsted trimmings with but little loss of its graphite 

Many mechanical lubricators employ a sight-feed arrange- 
ment through which the drops of oil rise through a sight glass 
filled with water; no difficulty is experienced with oil containing 
colloidal graphite, as the surface of the oil is not penetrated by 
the water. It is different with watery solutions of colloidal 
graphite such as aquadag; they obviously cannot be passed 
through water. Johnston has patented a lubricator with a sight- 
feed glass filled with kerosene, through which the drops of diluted 
aquadag sink down on account of their higher specific gravity 
as compared with kerosene. This arrangement has proved 
quite satisfactory for feeding aquadag into the steam pipes of 
engines using saturated steam. 

Drawbacks to the Use of Colloidal Solid Lubricants. One 
unsatisfactory feature of colloidal graphite solutions is their 
black, "inky" nature, which creates strong prejudice against 
their use on the part of operators of engines or machinery; 

1 ARCHBUTT and DEELBY, "Lubrication and Lubricants," p. 152. 


colloidal graphite stains are difficult to remove from the hands, 
etc. Colloidal talc will probably prove less objectionable in 
this respect than colloidal graphite. 

The great drawback to all colloidal solid lubricants is, however, 
their susceptibility to the action of electrolytes, e.g., acids and 
alkalies. The presence of electrolytes causes rapid destruction 
of the colloidal films and flocculation or separation of the solid 
lubricant from the liquid in which it is dispersed. The follow- 
ing experiments with dilute diffusions of oildag and aquadag in 
oil and water, respectively, containing various percentages of 
mineral acid, alkali, fatty acid, acetic acid, and petroleum acid 
show the tendency to flocculation. The oil used for the oildag 
experiments was a neutral filtered spindle oil to which was added 
the amount of oildag recommended by the makers, giving a 
graphite content of 0.35 per cent of the blended oil. The results 
are as follows: 

Mineral Acid. It was found that even the slightest trace of 
sulphuric acid (HoSO 4 ) precipitated the graphite. Flocculation 
within 24 hr. was caused by 0.1 per cent sulphuric acid; 0.005 per 
dent caused complete flocculation in 3 days. 

Alkali. The results with an alkali (caustic soda) were very 

Acheson himself has realized the importance of the purity 
of the mineral oils or water used for mixing with oildag or aqua- 
dag, respectively. He states: 

With de flocculated graphite the very best results will be obtained 
when the water or oil is absolutely pure, but commercially we may 
perhaps always have a very slight sedimentation of the graphite. The 
manufacture of practically pure or neutral petroleum oil may be made 
quite commercial, the presence of impurities in the oil now placed on 
the market being almost solely due to the failure of manufacturers 
properly to wash the oil. True, in some instances, while thorough 
washing may be performed with water, the water itself is not pure, 
which would still cause impurities to be found in the oil that would be 
capable of causing sedimentation of the graphite, but this residue, 
which is left by natural waters when they be of an impure nature, could 
finally be removed by a finishing wash with distilled water. 

It is a fact that most if not all acid-treated oils on the market 
are quite unsuitable for mixing with colloidal lubricants. The 


most suitable oils are undoubtedly these which during the process 
of refining have not been in contact with acids or alkalies but are 
refined by earth filtration only. Users of colloidal lubricants 
should therefore be warned not to mix them with the ordinary 
grades of oils, unless they have the assurance of their suppliers 
that none of the ingredients present contains acid or alkali or 
has been acid treated. 

Fatty Acids. The flocculating action of fatty acid is not so 
marked as with mineral acid. The graphite was precipitated by 
0.3 per cent of linseed-oil fatty acid in 4 days; 0.1 per cent of the 
same acid took 2 weeks to precipitate the graphite completely. 
Holde states that "free organic acid need not always act as a 
coagulant even with colloidal graphite; small quantities may 
under certain circumstances act as a stabilizer." 

This experiment shows that, if precipitation of the graphite is 
avoided, colloidal lubricants should not be mixed with fatty oils 
or compounded oils that contain a fair amount of fatty oil. 

Most oils used for marine steam engines, locomotives, and other 
severe services are heavily compounded with vegetable or animal 
oil (from 10 to 30 per cent) and contain an amount of free fatty 
acid, usually exceeding 0.5 per cent. 

Acetic Acid. The action of acetic acid was found to be similar 
in intensity to that of mineral acid. 

Petroleum Acids. Petroleum acids (of a fairly volatile organic 
character) may be produced, during use, in oils employed in 
circulation systems in automobile engines, gas engines, oil engines, 
and Diesel engines. 

In the author's experiments, petroleum acid was produced in 
the oil by blowing air through neutral filtered spindle oil heated 
to a high temperature (360 to 400F.) to accelerate the oxida- 
tion and the formation of acid. To the oil thus prepared was 
added the prescribed amount of oildag. The presence of 0.1 per 
cent of petroleum acid caused complete precipitation of the 
graphite in 5 hr. When the experiment was repeated with 
another sample of oil similarly treated, but only slightly " blown," 
containing 0.01 per cent of petroleum acid, the flocculating action 
was much less marked, but after 2 weeks complete separation 
took place. 

The amount of petroleum acid produced in the oil during pro- 
longed use in an automobile engine will not be very great. An 


average amount may be considered 0.01 per cent, assuming that 
the oil is a neutral filtered oil; and as oils for automobile use are 
fairly viscous, there is perhaps not much to fear from the presence 
of petroleum acid. Obviously, the more viscous the oil is the 
more slowly does the graphite separate out. 

Oils taken from enclosed high-speed gas and Diesel engines have 
been examined, containing over 3 per cent of free carbon in sus- 
pension, which had produced no ill effects on the engine. The 
carbon had been formed by carbonization of the lubricating oil 
inside the cylinders and had worked its way down into the 
crank chamber and mixed with the oil; probably a large amount 
of this carbon was present in colloidal form. It is a well-known 
fact that black waste oil from internal-combustion engines of all 
kinds cannot be freed from its carbon content by ordinary 
filtration and that gravity separation in settling tanks may take 
months to accomplish and is rarely completely satisfactory. 

The normal graphite content of 0.35 per cent in an oil blended 
with colloidal graphite, if separated out in an engine, would be 
considerably less than the 3 per cent of free carbon referred to 
above, but, its nature being different, only practical experience 
can determine the actual risk incurred, if any, by the use of such 
" impure" oils as will cause precipitation of the graphite. 

Emulsifying Effect of Water. A quantity of diluted oildag was 
mixed with an equal amount of distilled water and shaken in a 
reciprocating bottle-shaking machine for 5 min. at room tempera- 
ture. All of the colloidal graphite emulsified with the water and 
formed a tenacious sludge which on standing separated out 
between the clear oil at the top and the clear water at the bottom. 
It would appear, therefore, that colloidal lubricants should not 
be recommended for use in circulating-oil systems, when water 
is likely to enter the system, as is invariably the case with steam 
turbines; enclosed-type, force-feed lubricated steam engines; and 
the like. 

In many enclosed-type internal-combustion engines (auto- 
mobile engines, gas engines, etc.) there is no great likelihood of 
water mixing with the oil in service, and no objection can be 
raised to the use of colloidal lubricants from this point of view. 

When it is desired to apply colloidal lubricants temporarily 
to certain bearings, there is no objection to mixing them with the 
lubricant in use, independent of the character of the oil, because 


the mixture is immediately introduced, and there is no time for 
the colloid to separate out and cause trouble. 

If mixtures of "impure" oil and colloidal lubricants are used 
continuously for a period by one of the many comparatively 
slow-feed oiling arrangements (bottle oiler, siphon oiler, drop-feed 
oiler, pad oiler, etc.), the colloid will flocculate and accumulate, 
the flow of oil not being sufficient to wash it away. As a result, 
narrow oil passages are choked, the supply of lubricant ceases, and 
trouble may easily occur. 

Summary. Finely pulverized solid lubricants cannot be auto- 
matically used mixed with oil unless they are kept continuously 
mixed by a special stirring mechanism or by the motion of the 
parts to be lubricated. 

With colloidal lubricants, there is no difficulty in obtaining a 
perfect mixture, but it is imperative that only very pure oils be 
used for making the mixture, unless the conditions of service are 
such that flocculation of the colloid is not likely to lead to diffi- 
culties or trouble of a serious character. 



Bearings. Numerous experiences testify to the value of solid 
lubricants and of graphite in particular for use in bearings. 

One British railway reports that good results have been ob- 
tained by using either colloidal graphite or flake graphite mixed 
with their ordinary locomotive-engine oil. The graphite is not 
used for regular running (the compounded locomotive-engine oil 
would cause flocculation of colloidal graphite, and flake graphite 
cannot be suspended in the oil) but only as a temporary remedy, 
whenever important bearings are inclined to heat. 

Several works report that by continuous use of colloidal graph- 
ite mixed with pure mineral oils they have obtained excellent 
results on heavy-duty bearings (heavy pumping engines, etc.) 
which previously gave trouble, even when using oils heavily com- 
pounded with fixed oil. The bearings ran not only cooler but 
also with an appreciable reduction in consumption of oil and 
without flocculation of the graphite. 

Where no care has been taken to provide specially pure mineral 
oils, flocculation has occurred, and choking of oil channels, etc., 
has resulted. 


Some bearings of high-speed fans, which were troublesome 
with oil alone, ran reasonably cool when using the same oil 
mixed with colloidal graphite. 

A maker of dictating machines found that customers did not 
trouble to oil the motors; he tried oildag and found that, even 
when the motors received no oil for several months after the 
initial application of oildag, no scoring occurred, owing to the 
graphitized surfaces produced in the bearings. 

One maker of jaw crushers lubricated the Pitman bearing by a 
continuous flow of water mixed with some Hudson's-soap extract 
and bicarbonate of soda. The Pitman always groaned for about 
15 to 20 min. after starting up; after aquadag was used mixed 
with the water, the groaning entirely ceased. 

Saving in power has been reported by several firms resulting 
from the admixture of colloidal graphite with the oil in use. 

As the chief object in providing lubrication for ball and roller 
bearings is to maintain the highly polished hard surfaces in good 
condition, and little lubricating properties are required, it would 
appear inadvisable to use powdered or flaky solid lubricants for 
this purpose, as they would probably not improve the surface of 
the balls, rollers, or races; only colloidal lubricants seem to have a 
chance of success for such bearings. The only ball and roller 
bearings in which the nature of the lubricant has an influence on 
the friction are those in which pure rolling does not take place 
i.e., in three- or four-point contact ball bearings and in roller 
bearings that develop end thrust ; here some rubbing takes place 
under extreme pressures, and, if the surfaces are impregnated 
with an exceedingly fine solid lubricant, they are likely to operate 
with less wear and friction. 

Some large lifts have vibrator wheels about 5 ft. in diameter, 
which travel along a smooth shaft of 8 to 9^ in. diameter. 
These wheels are bushed with cast iron and require careful and 
reliable lubrication. It has been found that by replacing ordi- 
nary lubricating grease with a grease containing artificial amor- 
phous graphite the number of scored shafts and the amount of 
wear were materially reduced. 

The lubrication of worm and worm-wheel reduction gears is 
always difficult; the pressure between the teeth is very great; 
even with an abundant supply of oil, the friction consists of a 
certain amount of solid friction in addition to fluid friction. It 


is therefore to be anticipated that the use of graphite in connec- 
tion with the gear oil would prove beneficial, and the results of 
experiments carried out at the National Physical Laboratory 
Teddington, England, with oildag and flake graphite on Lan- 
chester's worm-gear testing machine show this to be the case. 
These experiments show that the addition of oildag to a mineral 
oil of relatively low oiliness improves the gear efficiency, so that 
the results are equal to those obtained by animal or vegetable 

Fine flake graphite (Foliac No. 100) also improved the effi- 
ciency with most of the mineral oils tested, and where an improve- 
ment was recorded it was greater than with oildag. The results 
appear, however, to be less consistent, and there was distinct 
evidence of greater wear than with oildag. 

When the temperature of the oil is increased, a critical point is 
reached above which the gear efficiency rapidly decreases. The 
effect of adding oildag or flake graphite was in every case to raise 
the critical temperature about 18C. so that an increased margin 
of safety in operation was thus obtained ; this happened even if the 
addition of solid lubricant did not increase the gear efficiency at 
lower temperatures. 

Steam Cylinders and Valves. In many steam plants great 
economies could be effected if the exhaust steam could be utilized 
for heating or drying purposes, for washing or cooking, or if 
the condensed steam could be used as hot feed water. One 
reason why this is not done more often is the presence of cylinder 
oil in the exhaust steam. The oil can be entirely eliminated from 
the condensed steam by electrical or chemical means but not 
from the exhaust steam itself before condensation, although good 
oil separators may take out as much as 99 per cent if the cylinder 
oil is pure mineral in character, i.e., not compounded with fatty oil,, 
such as tallow oil. 

The use of aqueous colloidal lubricants is probably limited to 
engines employing saturated steam and engines of small power; 
it must be kept in mind that if it were not for the water film 
produced by steam condensation in the cylinder, the friction 
would be very high indeed. In engines employing superheated 
steam, there is little or no condensation in the cylinders, and it 
becomes necessary to provide a lubricating film in order to avoid 
excessive friction and wear. 


Many experiments have been made with graphite and oil for 
the internal lubrication of steam engines employing superheated 
steam. Pure, fine flake graphite may be used, or colloidal 
graphite may be mixed with the cylinder oil, which should prefer- 
ably be a pure mineral oil, not compounded with fatty oil, as is 
the case with most good-quality steam-cylinder oils. The results 
of graphite employed in this way have in many cases- been very 
satisfactory; appreciable reductions in consumption of oil have 
been recorded, also less wear of internal moving parts. Alongside 
these results there are also a great many failures, although no 
failures have been reported with colloidal graphite. The failures 
with flake graphite have been due to excessive and injudicious use 
of the graphite or to the use of coarse or impure graphite or to 
breakdown of the complicated lubricators required to keep the 
graphite-oil mixture well stirred. 

Under superheat conditions the surfaces are more difficult to 
lubricate than with saturated steam, and the necessity for not 
overfeeding with graphite will be readily understood. Excess 
graphite accumulates behind the piston rings and in the metallic 
packing and will, in time, make the rings inflexible in their 
grooves, resulting in scoring of the surfaces, leakage of steam 
past pistons and piston rods, etc. Great care must be exercised 
in the use of flake graphite for superheated steam conditions, 
and only the purest graphite must be used, in order to avoid 
excessive wear of pistons, piston rings, cylinders, piston rods, 
metallic packings, etc. 

Graphite has been used by many marine engineers for lubricat- 
ing large, unbalanced D-type slide valves. Cast iron, being 
more or less. porous, is a material particularly likely to benefit 
from the use of graphite : when the pores are filled and a graphite 
coating is produced it will be found that an exceedingly small 
amount of graphite is required to maintain the surfaces in good 
condition. Impregnation of such surfaces with graphite reduces 
the tendency to abrasion and makes it easier for the cylinder oil 
to maintain efficient lubrication. 

When a mixture of flake graphite and cylinder oil is used for 
the internal lubrication of steam engines, the graphite will in time 
find its way out with the exhaust steam; it is easily separated 
from the steam and deposited in the oil separator or hot well. 


Flake graphite will adhere to the baffles in the separator, and 
accumulations should be removed at suitable intervals. 

In all cases where the use of graphite has brought about a 
reduction in consumption of steam-cylinder oil, it has also reduced 
the quantity of oil reaching the boilers and therefore reduced the 
possibility of boiler troubles from this source. 

Internal-combustion Engines. The use of solid lubricants for 
the internal lubrication of internal-combustion engines has been 
the subject of much controversy, and various opinions have been 
expressed regarding such features as preignition, carbon forma- 
tion, sooting of spark plugs, ease of starting, oil consumption, 
and reduction in friction. 

Preignition may be due to accumulation of carbon deposits, but 
the cause of preignition appears to be not so much the carbon 
itself as the earthy and other impurities (road dust, lime, iron 
oxides, etc.) which may be present in the deposit. Artificial 
graphite whether in the amorphous or colloidal form seems here 
to possess advantages over most natural graphite, as the presence 
of minute earthy impurities is more easily avoided in artificial 
graphite. As compared with the use of oil alone, the tendency to 
preignition may be said to be increased, more or less, according 
to the purity of the graphite. 

Carbon Formation and Sooting of Plugs. Contradictory reports 
are received with reference to this point. In small engines, such 
as motorcycles, no difficulty is experienced; some records even 
report less sooting of plugs when using colloidal graphite, but that 
may perhaps be explained by a more economical use of the oil 
when using graphite. 

The colloidal graphite is not consumed in the combustion space 
but, in the form of an exceedingly fine dust, spreads and adheres 
to the walls of the combustion chamber and the sparking plugs. 
The greater the consumption of oil the more graphite is deposited. 

The formation of oil carbon depends on the amount of oil 
burnt inside the cylinders and on the nature of the oil; some oils 
produce, more carbon than others, but the amount of oil carbon 
produced will normally never exceed 0.02 per cent of the oil 
used. Although hydrocarbon oils contain over 80 per cent 
of carbon, most of the oil is vaporized and decomposed into 
other hydrocarbons, with the result that the actual amount 


of oil carbon formed is a very small percentage of the amount of 
oil consumed. 

When mixing colloidal graphite with oil to give a graphite con- 
tent of, say, 0.35 per cent in the mixture, it must be remembered 
that this graphite is not consumed and that unless a large percent- 
age of the graphite is swept out by the exhaust gases, or a very 
large reduction in oil consumption takes place, the formation of 
carbon may easily be greater than with oil alone. 

Ease of Starting. Opinions appear to be unanimous that when 
graphite is used, engines (motorcycles, automobile engines, gas 
engines, etc.) start more easily and with greater freedom. The 
following contribution to Motor Cycling may be quoted : 

I found when using oildag that carbonization was markedly reduced, 
even under the very heavy lubrication that I give my engine as a rule. 
Engine "freeness" (allowing for the inherent freeness of the engine) is 
marked. The pressure of either valve spring acting on the engine 
via the tappet and cam was sufficient to rotate the back wheel of my 
T. T. single to the point of rest of the valve spring, so you can imagine 
there was not much friction in that engine. The cylinder walls took 
on a very high mirror-like polish. I found no concretion behind the 
piston rings, and what carbon deposit there was in the cylinder head 
and on the piston top was soft and easily removed. The effect of the 
graphite on the valve stems, particularly such a hot-working stem as 
that of the exhaust valve, was wonderful. The graphite was able 
to resist the heat and gave the valve stems a similar "mirror" surface 
to that of the cylinder. I further used it as a general lubricant for 
the cycle details. Carbureter slides polished and lubricated with oildag 
worked very smoothly and with a minimum of air leakages. It was 
also useful as a dressing for screw threads liable to bind or stick and on 
valve caps and plug threads; it made a good but easily broken joint. 

Oil Consumption. Saving in oil consumption, made possible 
by the use of graphite, is due to the smoother surfaces of pistons 
and cylinders and the more uniform and slightly smaller clearance 
space between them. Better compressions are also obtained due 
to less leakage past the piston. When the initial oil consumption 
is large, as with aircraft engines, the saving is apt to be overlooked ; 
but with small engines and where adjustable mechanical lubrica- 
tors are employed, the saving obtained may be quite considerable. 

Reduction in Friction. Speaking generally, half the friction in 
an internal-combustion engine is piston friction; the lubricating 


oil film is probably never complete, and so a certain amount of 
metallic contact (solid friction) invariably takes place. Porous 
cast-iron surfaces are easily filled and coated by graphite, and an 
appreciable reduction in friction may be anticipated when the 
necessary care is taken in the judicious use of the right kind of 

In the United States, colloidal graphite appears to be exten- 
sively used for the initial "running in" of automobile engines; 
it is said to save considerable time in producing a good surface 
and gives the engines a good internal skin before leaving the 
builders 7 works. 

Ropes, Chains, and Gears. Various greases are usually em- 
ployed for the lubrication and preservation- of ropes, chains, 
and gears, and, as already mentioned, the admixture of a small 
amount of good-quality finely divided graphite is beneficial. 
Messrs. Hans Renold, Ltd., London, find that with intermittently 
lubricated chain drives, graphite grease containing artificial amor- 
phous graphite is very suitable. When the chain has been soaked 
in the hot liquid grease it will work without further lubrication 
for a long period sometimes for several months whereas with 
thin oil in use it must be applied at least once a week, and then 
the results are not always satisfactory, a reddish deposit (rust) 
being found in the bearings of the chain. With graphite grease 
this deposit does not form. The same firm also reports that the 
clutch band in their power clutches, when lubricated by graphite 
grease, requires no attention for long periods. 

When chains or gears are enclosed in an oiltight casing, the 
use of an oil bath is preferable to grease ; in this case, the admix- 
ture of a small amount of finely pulverized graphite or colloidal 
graphite is also beneficial. 

Metal Cutting and Wiredrawing. Colloidal solid lubricants 
such as aquadag have been used as coolants for cutting 
purposes. Experiments seem to indicate that colloidal solid lubri- 
cants are not satisfactory when used for this purpose by them- 
selves. .They do not flow to the tool point if it is greasy, and the 
tool point therefore wears; when they are mixed with ordinary 
cutting emulsions or soap and water, good results have been 
obtained, but the high first cost of colloidal lubricants militates 
against their use for cutting purposes; the staining effect of the 
graphite on the hands of the operators is also objectionable. 


In metal wiredrawing operations semisolid lubricants are 
used, such as mixtures of olive soap and powdered talc. There 
appears to be no reason why a vegetable soap mixed with a 
suitable amount of finely powdered graphite should not be capa- 
ble of rendering good service. 

Aquadag is used in wiredrawing the metal filaments (Demp- 
sters Patent 17722, 1911) used in electric lamps; the dies require 
a certain amount of lubrication to produce a satisfactory thread, 
and aquadag is apparently the only nonoily lubricant that has 
given satisfaction for this purpose. 



Ring oiling is employed largely on modern high-speed shafting 
bearings, practically all electric motors and electric generators, 
also small steam turbines. Main bearings in most gas engines, 
Diesel engines, and horizontal oil engines, as well as many steam 
engines, employ the ring-oiling system for the crankshaft bearings. 

The advantages of ring oiling over the drop-oiling system are 

1. Better and more uniform lubrication. 

2. Greater oil economy. 

3. Greater factor of safety in operation. 

4. Greater cleanliness. 

5. Less attention required. 

Ring oiling is used for small as well as large bearings but not 
for the very smallest, say below 2 in., running at high speeds, 
as the rings frequently fail to operate, and it is difficult to prevent 
frothing and waste of oil. 

The bearing housing (Figs. 31 A and 3 IB) forms an oil reservoir 
in which the oil is maintained at a certain level, preferably indi- 
cated by an oil gauge, which may also serve for the introduction 
of the oil. 

On the shaft are usually suspended one or two rings or chains 
dipping into the oil ; when revolving, the rings carry oil to the top 
of the shaft, from which it runs into the oil-distributing groove on 
the " on side" of the journal and spreads over the bearing sur- 
faces. This groove extends to within J^ in. of the bearing ends 
and is well chamfered to facilitate the oil's wedging its way 
into the bearing. If the motor is reversible, two oil-distributing 
grooves are obviously needed, one on each side. No other oil 
grooves should ordinarily be made, as the high surface speed 
assists the oil materially in getting in between the rubbing 

When the surface speed is low, and the bearing pressure high, 
oil grooves may be advantageous, as explained on page 121. 




When speeds are low, or the oil becomes thick (cold mornings), 
oil rings may not start readily. Figure 29 illustrates an oil ring 
with notches filed on the inside, which are said to assist the ring 
in starting, owing to the greater friction between the ring and 
the shaft. 

For low-speed bearings, oil rings are sometimes replaced by 
chains, which touch the shaft over a long arc and are therefore 
kept in motion with greater certainty than plain rings. At high 
speeds, chains have the disadvantage that the links churn the oil, 
which leads to foaming and leakage of oil from the bearings. 

When the speeds are exceptionally low, say a few revolutions 
per minute, neither rings nor chains are satisfactory, and oil 

12345 6 

FIG. 29. Notched oil ring. 

FIG. 30. Various types of oil rings. 

rings or collars fixed on the shaft are employed, whence the oil is 
removed by stationary scrapers in the upper part of the bearing, 
guiding the oil afterward into the oil-distributing groove. Such 
bearings are employed on the drying cylinders of papermaking 
machines and other slow-speed machinery but are also occa- 
sionally used for moderate-speed machines. Oil rings are usually 
made twice the diameter of the journal, the oil level being at 
a distance below the shaft about half its diameter. 

One oil ring will suffice for bearings up to 8 in. in length. Two 
oil rings are needed for bearings from 8 to 16 in. in length; and 
three for larger bearings. 

As to the shapes of oil rings, there are a great many; some are 
indicated in Fig. 30. They should preferably be made undivided. 
When made in two halves and jointed, slight wear may cause 
them to operate irregularly or even refuse to revolve. Uneven- 
ness at the joint at high speed leads to foaming of the oil and oil 



spray. Unevenness or roughness on the two sides of the oil ring 
will have similar effects. Rings that are slightly oval, due to 
lack of care during manufacture, will obviously be inclined to 

In large bearings, cooling of the oil by introducing a cold-water 
coil in the reservoir may be found desirable or even necessary 
under severe conditions (see ' 'Turbines ") 

Difficulties sometimes occurring with ring oiling are 

1. Foaming and spraying. 

2. Leakage, endways or sideways. 

Foaming and Spraying. Troubles of this kind may be due to 
too high revolving speed of the oil rings or to too low an oil 

FIG. 31. Ring-oiling bearing. 

level (which brings about quick speed of the rings owing to the 
smaller resistance offered to the movement of the rings through 
the oil). The oil is violently thrown away from the rings, form- 
ing oil spray, oil foam being formed by the rings' drawing air 
into the oil where they enter the surface of the oil. Excessive 
foaming and oil spray always mean waste of oil, as the finest oil 
spray finds its way through the bearing ends or covers. Such 
loss of oil may become dangerous by lowering the oil level so 
much that the bearings will receive too little oil. 


Leakage through the side of the bearing between the top and 
bottom parts can be overcome by inserting a thin leaden wire 
which, when the bearing is put together, is squeezed flat and forms 
a seal (Fig. 31D). The lip (1, Fig. 315) is intended to prevent 
such oil leakage; also the longitudinal drainage groove (2, 
Fig. 3 1C), which is sometimes found in large bearings, draining 
the oil back to the reservoir at its two ends. 

The proper remedy is, however, to have a type of oil ring suit- 
able for the size and speed of the shaft. The speed of the oil 
rings is governed by the propelling force caused by oil adhesion 
between the ring and the journal where they touch at the top 
and the retarding force caused by the opposition to movement 
created by the speed of the ring through the oil in the well. It 
will be recognized that oil rings 2, 3, 4, and 5 will give lower 
ring speeds (more slip) than oil ring 1, whereas No. 6 will usually 
give greater speed than No. 1 owing to less surface in contact 
with the oil therefore less resistance with the same propelling 
force. By finding out from practical experiments which type 
of ring gives a suitable speed and oil feed, the majority of troubles 
with ring-oiling bearings may be overcome. 

It is particularly important to study this point when outside 
the bearing there is a pulley, which in revolving creates a suction, 
tending to increase loss of oil from the bearing. The oil when 
leaving the ends of the bearing drops back into the oil reservoir 
and is thus kept in constant circulation. If the bearing is well 
designed, there will be very little oil waste by leakage or by oil's 
creeping along the shaft. Sometimes the oil creeps spirally along 
the shaft and is thrown away where it is least desired, creating 
unsightly oily floors or spoiling fabrics, e.g., cloth looms. The 
remedy may lie in lowering the oil level or altering the type of 
oil ring, but the root of the trouble may be wrong shaping of 
the bearing surfaces. Figure 31 A shows how the edge must be 
rounded off, which helps to prevent oil creeping, whereas a sharp 
edge does not hinder the oil in passing along the shaft. 

An excellent arrangement is to have a circumferential oil groove 
with sharp edges, as shown in Fig. 31 A, with a drainage hole at the 
'bottom; in addition, a longitudinal drainage groove also with 
sharp edges will assist in preventing excess oil's reaching the bear- 
ing ends, as shown at the right-hand side of Fig. 31-4. Other 
methods rely upon oil throwers formed on the shaft and suitable 


shapes of bearing housings at the end to receive the oil and convey 
it to the oil reservoir. 

Figure 32 shows the simplest form of oil throwers. This con- 
struction, with a drainage passage from left to right through a 
center wall, in one case caused the oil to overflow from the left- 
hand chamber, owing to the passage's being almost choked with 

Bearing Clearance. A clearance of 0.002 + 0.001 in. per 
inch of shaft diameter represents normal practice and will give sat- 
isfaction as long as the deflection of the shaft due to an overhang- 

1 011 

2 OH Throwers 

FIG. 32. Simple oil thrower. 

ing pulley, heavy flywheel, or rotor close to the bearing does not 
exceed this clearance. 

To take care of such shaft deflection, medium- and large-size 
bearings are frequently self -aligning, being made with spherically 
seated housings. 

Care of Ring-oiling Bearings. When ring-oiling bearings are 
well designed, with large oil wells and employing good-quality 
oils, they will operate for long periods without undue attention 
or loss of oil. During the first few weeks of service the oil wells 
of new bearings should be emptied and recharged with fresh oil 
every few days in order to remove any molder's sand or grit 
which may still be present in the bearing. Once the bearing is 
clean, and a good skin formed, it will be sufficient to empty the 
wells every 3 months and recharge them with fresh oil or a 
mixture of filtered and fresh oil. When the bearings are situated 
in dusty surroundings, more frequent changes of the oil may be 
desirable, as dust will find its way into it, notwithstanding 
precautions in the way of wooden or felt rings fitted in the bearing 
ends; of course, such rings do reduce the amount of dust that 


enters. Where chemical fumes are present in the air, which have 
a destructive effect on the oil, it must also be changed frequently. 

When good-quality mineral oil is used, it can be filtered and 
used over and over again, so that the oil consumption per year is 
usually very small. When oils with a mineral base but com- 
pounded with animal or vegetable oils are employed, they will 
develop gumminess in the bearings and necessitate cleaning. 
Such cleaning is unnecessary when straight mineral oils are 
employed; cases have been known where good-quality oils have 
been in use for years without any real necessity for cleaning the 
bearings or the oil wells. 

When a change is made from a compounded oil to a mineral oil, 
the latter will loosen the accumulations formed by the old oil. 
In such cases it is advisable to renew the charge after a few weeks' 
run; the oil when withdrawn will be very dark in color, owing to 
the deposits and perhaps very slight initial wear due to the bearing 
surfaces' adapting themselves to the new oil, but a fresh charge 
ought to work clean, if the new oil is of the right quality and the 
bearing has been completely cleansed. 



Satisfactory bearing lubrication, i.e., cool running and inappre- 
ciable wear, is very important. If wear takes place, the rotor 
is lowered, and the magnets will then exert a pull on the rotor in 
a downward direction, which further increases the bearing pres- 
sure and accordingly the wear. Most bearings therefore operate 
with bearing pressures of 50 to 100 Ib. per square inch and are 
oil flooded. 

Ring-oiling bearings are most frequently used. 

Ball bearings are also coming much into use, particularly as 
smaller size horizontal bearings and as vertical bearings and in 
dusty surroundings, in which case grease is often used in prefer- 
ence to oil. 

When a new generator or motor exhibits a tendency to develop 
heat in one bearing, the bearing should, of course, be examined. 
If it be found in good condition, the cause of the heating may be 
found in the thrust of the armature shaft against the bearing, 
which may result from one of two conditions. First, the machine 
may not be level, and the armature shaft may "dip." Second, 
the magnetic centers of the pole pieces and armature may not be 
in line; i.e., the pole pieces may not be exactly centered in their 
relation to the magnetic center of the armature axially, and, as 
the tendency of the armature is to run to the true magnetic 
center, it will automatically tend to move toward that position, 
which may cause the shaft collar to rub against the bearing at one 
end and cause heating. 

The unbalanced magnetic condition may have been caused by 
forcing the armature not quite far enough or a trifle too far on to 
the shaft in the factory. Some motors are furnished with slots 
in the field-magnet yoke through which the field-magnet cores 
are bolted to the yoke, and the cores may be shifted a trifle to 
the right or left to compensate for any slight axial unbalancing 
of the magnetic center as compared with that of the armature. 




Moving the magnet cores KG in. will, as a rule, be sufficient to 
give relief. If the field-magnet yoke is not slotted, a light cut 
may be taken off the shoulder of the shaft, at the end that rubs 
against the bearing, in order to obtain the necessary clearance. 

Oil throwing from the bearings into the generator or motor is a 
troublesome disease, often very difficult to cure. Several reme- 
dies have been mentioned under ring-oiling bearings, but they are 
not always effective with those types of electric dynamos which 
create a draught. When the generator is enclosed, and the venti- 

FIG. 33. FIG. 34. 

FIGS. 33-35. Oil throwers. 

FIG. 35. 

lation led in from below, this danger does not exist or, at any 
rate, exists to only a slight extent; but when the generator is not 
enclosed, the oil finds its way into the ventilating ducts, tending 
to choke them with dirt and dust, which adhere to the oil. The 
oil also gets on to the commutator or slip rings and causes 

In Figs. 33, 34, and 35 are illustrated some more elaborate 
methods adopted to prevent oil throwing. Figure 34 shows a 
series of thin copper plates, which only lightly touch the shaft 
and thus create an effective labyrinth seal; but they are inclined 
to cause wear of the shaft, small gritty particles embedding them- 
selves in the soft copper surfaces. Certain information on this 
subject is also given under " Turbines.'' 

It is said, and it sounds quite feasible, that compounded oils 
do more damage when getting on to the windings than straight 
mineral oils, as they absorb moisture and thus reduce the insula- 


tion resistance of the armature more than straight mineral oils, 
which are not hygroscopic. 

Commutator Lubrication. Lamp oil (kerosene) used very 
sparingly is probably the best oil to keep the commutators clean 
and well lubricated. It also softens the mica and thus causes 
it to wear down so that it does not stand out beyond the bars. 


Three grades of dynamo oils will take care of most require- 
ments; they are all pure mineral oils, viz., bearing oils 2, 4, and 
5 (see page 135). The oil is somewhat exposed to oxidation dur- 
ing its continuous circulation in the bearings, but as long as the 
bearing temperatures do not exceed, say, 120F., there is no 
need to have specially prepared dynamo oils; where the oil 
temperatures exceed 120F., circulation oils of the corresponding 
viscosities should be preferred to ordinary bearing oils. 

A rough guide for selecting the correct viscosity of dynamo oil 
is given in the following chart: 

For Electric Generators and Motors 

Bearing Oil 2* or Circulation Oil l.f For small electric generators or 
motors up to 50 hp. and up to 100 hp. when there is no excessive belt pull 
on the shaft close to the bearing. 

Bearing Oil 4 or Circulation Oil 2. For larger electric generators and 
motors under normal operating conditions and for motors below 100 hp. 
with excessive bearing pressures. 

Bearing Oil 5 or Circulation Oil 3. For generators or motors above 100 hp. 
operating with excessive bearing pressures. 

Ball-bearing grease is employed only for smaller motors, operating in 
dusty surroundings or in hot and moist climates. 

* See p. 57. 
t See p. 243. 



Horizontal thrust bearings are designed to take up axial thrusts 
of revolving parts, e.g., in horizontal centrifugal pumps or turbines 
or the propeller shafts in marine steam engines. 

Vertical thrust bearings are employed to carry the weight of 
revolving parts, e.g., in vertical water turbines or centrifugal 
pumps and vertical electric generators or motors. 

,. ....-_ T .-._._, __ r _ _ --:. - _- : , _ '-sf^y^s^ 

FIG. 36. Ring-oiled thrust bearing. 

Collars on the shaft transmit the pressure to stationary collars, 
various means being employed to introduce an oil film between 
the rubbing surfaces, as described under " Turbines " and 
" Marine Steam Engines." Figure 36 illustrates a method of 
oiling the thrust bearing in high-speed pumps. By means of an 
oil ring (1) the oil is thrown off the collar (2) against the oil 




catcher (3), whence it runs into the oil cup (4) and reaches the 
hollow shaft, finally returning to the oil well. The lower part of 
the bearing is water cooled. Figure 37 illustrates an unsatis- 
factory method, as the large oil disk (2) causes foaming and 
creates heat. 

Figure 38 illustrates an ingenious method of providing oil 
circulation in a vertical thrust bearing supporting a shaft revolv- 

FIG. 37. 

FIG. 38. Vertical thrust bearing automatic oil 

ing at 1,500 r.p.m. upon which is fitted an electric motor driving a 
centrifugal deep- well pump at the lower end of the shaft. 

The shallow spiral grooves on the part (1) lift the oil into the 
oil chamber (2); the oil pressure created here drives the oil 
through the oil drillings in the shaft; the oil after doing its work 
reaches the oil-return channel (3) and the oil well (7) w;hich is so 
arranged that the oil cannot overflow down the shaft. The drain 
plug (4) is removed when it is desired to empty the bearing, and, 
when the bearing is being filled through the filling plug (5), the 
overflow plug (6) is removed so as to ensure a correct oil level. 



When thrust bearings have only one collar or one rubbing surface, 
they are called step bearings or pivot bearings. 

Figure 39 illustrates the simplest form of step bearing; the 
revolving shaft rests on three washers; the top washer may be 
arranged to revolve always with the shaft, so as to save wear of 
the shaft itself. With low bearing pressure only one washer is 
needed; the higher the pressure the more washers are required. 
When one washer begins to heat and seize, it stops revolving, and 
one of the cooler washers starts to 
revolve, so that they divide the work 
between them, only one washer act- 
ing at a time. 

The washers should always have 
shallow radial oil grooves cut in 

FIG. 39. Step bearing. 

FIG. 40. Water turbine bearing. 

their rubbing faces, the grooves stopping slightly short of the 
edges; and the trailing edges of the grooves should be well 
rounded to facilitate the entrance of oil between the surfaces. 

The oil enters the central hole; rises to the top owing to cen- 
trifugal action; and returns through the drain hole. When such 
bearings get uncomfortably warm, a remedy is to increase the 
flow of oil by means of a pump which forces the oil in under 
pressure; the oil returns from the top through a pipe into an oil 
reservoir, whence the pump draws its supply. In this way a 



FIG. 41. 

FIG. 42. 
FIGS. 41-42. Water turbine bearings. 


greater radiating surface is obtained, and the bearing will run 

Vertical water turbines make use of bearings as illustrated in 
Figs. 40, 41, and 42. In Fig. 40 the weight of the turbine is taken 
by a stationary vertical shaft (1) which has an oil reservoir (2) 
at the top with a bronze washer (3). The hollow revolving 
shaft (4) carries the weight of the revolving parts and transmits 
the pressure through the hardened-steel part (5) to the washer 
(3). Oil is fed through the top from a sight-feed drop oiler (6). 
The overflow oil runs down into the guide bearing (7), which is 
under water; at the lower end there is a gland packing to prevent 
entrance of water into the hollow shaft, but, as some water gen- 
erally gets in, compounded oils, such as marine-engine oil 1 or 
2 (see page 267) should be used, which will emulsify with the 
water and maintain efficient lubrication. 

In Fig. 41 the shaft (1) and washer (2) revolve; the stationary 
washer (3) receives the full pressure and transmits it through the 
stationary shaft (4) and cover piece (5) to the casing (6). The 
oil circulates continuously through the oil grooves, which extend 
right to the edge. It will be noticed that the bearing surface 
in this design is very small; the bearing is very compact, so that 
a rich viscous oil, such as marine-engine oil 1, must be used. 

. In Fig. 42 the bearing surface is much bigger, also the radiating 
surface is greater, which gives cooler running; ordinarily, a slow 
oil feed from a sight-feed drop oiler suffices; but where high 
pressure exists and the heat developed is great, an oil-circulation 
system may be employed. 

The oil to use in such bearings as Figs. 37, 38, and 42 may well 
be circulation oil 1 or 2, as the bearing pressure is low, and 
the revolutions are usually high, say above 100 r.p.m. 

A special type of step bearing is employed in the Curtiss vertical 
turbines, as described on page 238. 

By far the most satisfactory and reliable type of thrust bearing 
for heavy pressures, whether the speeds are high or low, is the 
Michell or Nomy single-collar thrust bearing. The collar or foot- 
step rests upon a fixed bearing surface, which is divided into a num- 
ber of segmental pads, each pivoted so that it is free to rock and 
take up any inclination to the moving surface that the conditions 
of speed, pressure, and viscosity of the oil may demand. Figure 
43 shows two Michell methods of supporting the pads, viz., pivot- 



ing along a line and pivoting on a point, the pivoting line or point 
being placed a little behind the center of each pad; experience 
shows this to be important to give a perfect film and therefore 
minimum friction. As the collar (1) moves over the pad (2), a 
wedge-shaped oil film is established; the oil is continuously drawn 
in at the leading edge, where the oil film is thickest, and escapes at 
the trailing edge, where the oil film is exceedingly thin. Some oil 
also escapes along the sides of the pad, as indicated in Fig. 44, 
which shows the directions of oil flow. These interesting photo- 
graphs are reproduced by the courtesy of W. J. Hamilton Gibson, 

FIG. 43. Michell thrust pads. 

more details being given in his paper before the Institution of 
Naval Architects, London, April, 1919. 

In some interesting experiments made by Brown, Beven & 
Co. the effect of slightly rounding the leading edge of the pads 
was found to be an increased carrying power and a slight shifting 
further aft of the center of pressure; these experiments also 
confirmed MichelPs opinion as regards the shape of the pads, 
which should be approximately square to give the best results. 

The pads are usually white metaled, and one might ask why 
go to this trouble, as there is no metallic contact? The character 
of the metal of the lubricated surfaces ought not to influence the 
results, as long as they are strong enough to stand the pressure; 
fine particles of grit or dirt may, however, be carried in between 
the surfaces with the oil; in that case the white metal will become 
abraded, and this is preferable to injuring the collar, which would 
occur were the pads made of hard material. 

The oil film is very thin sometimes less than 0.001 in. so that 
the bearing surfaces must be carefully scraped, and oil grooves 



Fia. 44. 



must on no account be cut, as they will allow the oil to escape 
and prevent proper film formation. 

The heat generated in the bearing is entirely due to internal 
fluid friction in the oil film, there being no metallic contact; 
the frictional heat is therefore dependent upon the area of the 
thrust pads, the rubbing speed, and the viscosity of the lubricant. 
The makers supply particulars as to the amount of heat generated 
under specific conditions and the quantity of oil and cooling water 
required to give the best results. 

The number of pads may vary but is usually six. They may 
be arranged in the form of an inverted horseshoe, as in the self- 

FIG. 45. Michell marine thrust bearing. 

contained marine thrust bearing (Fig. 45), suitable for both geared 
turbines and marine steam engines; or evenly distributed over 
the collar, as in the geared turbine thrust bearing (Fig. 46). 

In Fig. 45 the collar bears against two inverted horseshoe- 
shaped surfaces (one for ahead and one for astern thrust). Each 
of these surfaces is subdivided into six pads pivoted on the ends of 
a corresponding number of screws. The shaft is supported by 
two ordinary journal bearings, and the well in which the collar 
revolves is filled about half full of oil, which lubricates the blocks; 
the journal bearings have upper keeps, fitted with siphon oilers, 
and a light sheet-iron cover forms a dust shield. 

The housing consists of one main casting and is water-jacketed 
in large-size bearings or when the speed is high. 

In Fig. 46 the shaft is carried in two journal bearings, the same 
as in Fig. 45, but the housing is made in halves, and the blocks 
instead of being independently adjusted are mounted in spherical 



seats and adjust themselves automatically. This type of bearing 
is not self-contained, as in Fig. 45, but must be connected to an 
oil-circulation system, usually a branch from the main-turbine 
oiling system. The blocks may be either "line pivoted " on 
the spherical seats or " point pivoted/' as shown. 

For steam turbines, where the thrust bearing is combined with 
the main bearing at the high-pressure end, and when the thrust 
does not exceed 5,000 lb., a much simpler form of Michell 

FIG. 46. Michell turbine thrust bearing. 

bearing is designed, one type having only one pivoted pad on 
either side of the collar. The Michell thrust bearing is also 
used with great success as vertical thrust bearings, required 
for vertical water turbines, centrifugal pumps, vertical electric 
generators, etc. 

It will be recognized that a perfect oil film cannot be established 
in the ordinary form of thrust bearing in which the coefficient of 
friction is about 0.03, whereas in the Michell bearing it falls to 
0.002 or even less. The Michell thrust bearings will safely carry 
a load of 400 to 500 lb. per square inch with rubbing speeds from 
60 to 100 ft. per second, without danger of metallic contact. 
Michell thrust bearings have run with no abnormal heat, carry- 
ing a pressure of 5 tons per square inch, a pressure at which the 
white metal surfaces of the pads began to flow like butter, thus 



showing that with perfect film formation the oil film will stand 
enormous pressures. 

The Kingsbury thrust bearing is designed on very much the 
same lines as the Michell bearing. Kingsbury also divides the 
supporting surface in segmental pads, and the pads are made self- 
adjustable by rounding the supporting surface, as shown in 
Fig. 47. The pad (2) rocks over the support (3) and thus allows 
a wedge-shaped perfect oil film to form between the pad and the 
revolving collar (1). 

The oils used for Michell thrust bearings are mineral oils of 
suitable viscosity. There is no need for compounded oils, as 
the film formation is not influenced by the oiliness of the oil. 
All that is required is viscosity. For slow-speed conditions, 
viscous oils are required, like bearing oil 4 (see page 135) or 

FIG. 47. Kingsbury thrust pad. 

circulation oil 3 (see page 243); for high-speed conditions, as in 
steam turbines, the same turbine oil is employed for the Michell 
thrust as for the turbine bearings. The Nomy bearing principle 
is somewhat similar to the Michell principle and is further 
explained in Chap. XIII, page 195. Nomy thrust bearings 
therefore offer the same advantages as Michell thrust bearings 
as compared with ordinary thrust bearings. 

Compounded marine-engine oils may be used for steam-engine 
Michell or Nomy thrust bearings for the sake of simplicity, but 
straight mineral oils operate cleaner and should be preferred. 


Ball and roller bearings operate on different principles from 
plain bearings ; the rolling contact between the balls or rollers and 
the stationary or revolving surfaces (ball races, roller races) 
is theoretically only point contact in ball bearings, and line 
contact in roller bearings, whereas ordinary bearings have 
large surfaces in rubbing contact at all times. When machinery 
equipped with ordinary bearings is started the frictional resist- 
ance is great, several times as great as the resistance after a 
couple of revolutions, when the oil film has been established; 
whereas in ball and roller bearings the friction at starting is the 
same as or only very little more than the friction during opera- 
tion, and is always lower than in plain bearings. 

It is this great advantage that ball and roller bearings have 
over plain bearings which is chiefly responsible for their ever 
widening use, particularly in machinery that frequently starts 
and stops or changes its direction of rotation, such as auto- 
mobiles, motorcycles, bicycles, reversible electric motors, 
railway turntables, etc. 

Roller bearings may possibly stand rough use, vibration, and 
shocks better than ball bearings, but they will not carry heavier 
loads, as many people seem to think, and at very high speeds 
ball bearings are usually preferred. Professor Goodman has 
made a lengthy study of ball and roller bearings, and the fol- 
lowing remarks are largely based on the information given by him 
in papers read before the Institution of Civil Engineers, 1911- 
1912, and the Institution of Automobile Engineers, in 1913. 


The rollers are nearly always plain cylindrical. Most bearings 
have a cage, as in Fig. 48, to hold the rollers in position. 

The Hyatt bearings (Fig. 49) have rollers which are helical 
springs, alternately of right- and left-hand pitch, and are much 
used for line shafting. 




The Timken roller bearing (Fig. 50) has two rows of tapered 
rollers and is used largely for automobiles. 

The pressure on the narrow line of contact between roller and 
shaft is great; hence, soft materials are liable to be crushed, 
and the wear is excessive. For high pressures the rollers, sleeve, 

FIG. 48. Roller-bearing cage. Fio. 49. Hyatt rollers. 

and casing liner should be steel hardened and ground so as to 
minimize the wear. 

During operation the roller cage moves at approximately half 
the journal speed, and the rollers revolve at very high speed, 

Fio. 50. Timken roller bearing. 


rubbing with their ends against the cage, so that these points 
require lubrication more especially because the rollers often 


create considerable end thrust. As such end thrust forces the 
cage against the inside of the bearing housing, lubrication is 
also required for these additional rubbing surfaces. 

When the rollers are not absolutely parallel with the shaft, or if 
they are the least bit taper, or if the shaft or sleeve against which 
they revolve is taper, the rollers tend to roll in a helical path. 
They push themselves against one end of the casing until the 
pressure becomes sufficiently great; then they slip back suddenly 
and start rolling afresh in a helical path toward the same end of 
the casing as before. The amount of end thrust created is largely 
dependent upon the bearing load (it may be as high as 30 per cent 
of the load) and does not appear to vary with the amount that the 
rollers are out of truth. 

The rollers have been known to wear right through their casing 
and nearly through the housing itself. 

With change in direction of rotation the end thrust is always 
reversed. Speaking generally, the starting effort of roller bear- 
ings is only slightly greater than the running effort, but when 
there is considerable end thrust the starting effort may be even 
twice as great as the running effort. 

The main evils of end thrust in roller bearings are: 

1. It is largely the cause of the f fictional resistance. 

2. It causes excessive wear on rollers, cage, shaft, and casing. 

3. It causes the bearing to run hot. 

4. It sets up vibration and rumbling in the bearing and its housing. 

The makers of the Hyatt roller bearings claim that one-half 
of the helical rollers will tend to run toward one end of the casing 
and the other half toward the other end and that end thrust 
is therefore eliminated. 

Speaking generally, roller bearings even the simplest types 
develop less friction than plain bearings, provided, of course, 
that they are erected with a reasonable amount of care. Bearing 
housings should be self -aligning, so as not to set up undue stresses 
anywhere in the bearings. 

To insure the best results, both ball and roller bearings must be 
very accurately fitted and, if worn, must be renewed and not 
allowed to run. If they are slightly out of line or slightly worn, 
great stresses are set up; the friction is high may even be higher 
than with plain bearings and the balls or rollers may break. 


The coefficient of friction of roller bearings is always higher for 
small than for high loads and considerably increased when there 
is appreciable end thrust. It ranges from 0.002 to 0.007. The 
normal average values for the coefficient of friction may be taken 
as 0.003 to 0.004; but for bearings of the Hyatt type, the values 
are higher, ranging from 0.0045 to 0.007, the lower values corre- 
sponding to high loading. 

Goodman summarizes the general results of his tests of roller 
bearings as follows : 

1. The coefficient of friction of roller bearings is greater at low than 
at high loads, but it is much more nearly constant than it is in plain, 
lubricated bearings. 

2. The coefficient of friction of roller bearings in which there is pure 
rolling is very nearly constant at all speeds; but when there is end 
thrust, the friction decreases as the speed increases. 

3. The coefficient of friction is independent of the temperature of 
the bearings unless the end thrust is excessive. 

4. The starting effort is very little greater than the running effort. 

5. The friction in a well-designed bearing is not greatly affected by 

6. The wear of the rollers is often excessive if the whole of the rotating 
parts and the casing are not hardened and well finished, especially when 
the bearing shows end thrust. 

7. The end thrust on the rollers varies almost directly as the load 
on the bearing and usually diminishes as the speed increases. The 
direction of the thrust is usually reversed when the direction of rotation 
is reversed. 

8. Other things being equal, the frictional resistance of bearings 
fitted with large rollers is less than with small rollers. 

9. The safe load for a given bearing diminishes as the speed of rotation 
of the rollers increases. 


Ball bearings cannot create end thrust; herein lies one of their 
great advantages over roller bearings, particularly at high speeds. 
They are less inclined to heat than roller bearings, as the friction 
is lower. The starting effort is the same as the running effort, 
and, in consequence, ball bearings, notwithstanding that they 
have only point contact as compared with line contact in roller 
bearings, are able to sustain as heavy loads as roller bearings. 



Ball bearings must not be adjustable; once the bearing is 
assembled, all running clearances must be correct, and neither 
the balls nor the races must wear. 

FIG. 51. Four-point con- FIG. 52. Three-point con- FIG. 53. Two-point 
tact ball bearing. tact ball bearing. contact ball bearing. 

Four-point and three-point contact bearings (Figs. 51 and 52) 
are not so satisfactory as two-point contact bearings (Fig. 53) for 
the reason that there is a grinding action between the balls and 
the races, and the balls get scratched. Two- 
point contact bearings may have flat races, as 
Fig. 53, and the results are very satisfactory; 
in fact, the coefficient of friction is lower than 
in other types of bearings, but the load- 
carrying capacity is 2 to 2^ times greater 
with grooved races, as in Fig. 54. With flat 
races the coefficient of friction decreases with 
increase in load, but with grooved races it 
may increase, possibly owing to the increased 
area of metallic contact between the balls and 
the races. For heavy loads the grooved races 
are to be preferred, given good alignment and 
workmanship; but if there is any doubt as to 
these points, flat (or cylindrical) races may 
prove better, as a slight lack of alignment will 
not affect the balls on a flat surface but may 
cause them to jamb and get cracked when 
running in grooved races. 
Goodman has found that the friction in ball bearings is never 
reduced by lubrication but is sometimes greater than when the 
bearings run dry. Bearings with flat races have run dry for long 

FIG. 54. Ball 
bearing with grooved 



periods without any apparent ill effects; but bearings with grooved 
races soon begin to whistle and grind, probably because there is 
more actual rubbing between the balls and the grooves than with 
flat races. As, however, the absence of lubricant means that the 
surfaces in time will rust, which is fatal, lubrication is always 

It is important that the balls shall be all of the same size; if 
some of the balls are smaller than others, the big balls have to 
take more than their share of the load (being bigger, they get 
more squeezed than the little ones); the smaller balls take less 
than their share, therefore slip more, and it is 
this slipping that causes the balls to deterioate 
and get scratched. 

All the best makers will guarantee first-class 
balls to be accurate within 0.0001 in..; it does 
not matter much whether 1-in. balls are slightly 
more or less than 1 in. in diameter, but they 
must all be exactly alike, and with properly 
made bearings the wear will then be practically 

When balls are overloaded they become 
covered with tiny flakes of "snow," the flakes 
being tiny crystals which have broken away 
from the surface of the ball; these specks can 
be seen only under the microscope with 300 to 
400 diameters magnification. When a ball 
finally breaks down it peals on one hemisphere FIG. 55. skefko 
and, curiously enough, usually only on the one 8Wlve eanng * 

The question of alignment of ball bearings is as important as 
in the case of roller bearings, if not more so. The bearing hous- 
ings are therefore usually made self-aligning, but in one type of 
bearing the Skefko the spherical outer ball race (Fig. 55) 
allows the inner race and balls to swivel out of their plane of 
rotation, so that they can adjust themselves to any lack of align- 
ment of the shafting, whether due to bending or due to bad 
erection, and the adjustment will of course take place with much 
greater ease than in the case of a self-aligning bearing housing. 

This bearing has other features. As there are two rows of balls, 
the load is distributed at any instant over three balls instead of 



on one or two, as in an ordinary ball bearing; this feature increases 
the load-carrying capacity. The bearing is also capable of taking 
a certain amount of end thrust, as the balls touch the spherically 
shaped outer race at points where the pressure between them is 
at a slight angle with the vertical plane. 

The inner race of a ball bearing must not be slack on the shaft; 
the shaft should preferably be ground to a light tapping fit for 

FIG. 56. Ball bearings for electric motor. 

the inner race, and the bearing secured against a shoulder by 
means of a nut (see Fig. 56). The outer race must not be a 
driving fit in the housing but should have an easy sliding fit, 
as otherwise the balls will be unevenly loaded. 

Figure 56 shows the correct method of mounting ball bearings 
on an electric motor; the right-hand outer race is not allowed 
much movement between the housing covers, but the left-hand 
one has freedom to slide in its housing when the shaft expands 
or contracts. 

The coefficient of friction of ball bearings is always greater with 
small than with high loads; it ranges from 0.001 to 0.003, the nor- 
mal average value being 0.0015 to 0.002. 

Goodman summarizes the results of his tests of ball bearings 
as follows, and his interesting remarks concerning a comparison 
between ball and roller bearings are also quoted : 


1. The coefficient of friction of ball bearings with flat races decreases, 
and with grooved races sometimes increases, as the load is increased; 
but it is much more constant than that of plain, lubricated bearings. 


2. The coefficient of friction of ball bearings is practically constant, 
at all speeds but has a slight tendency to decrease as the speed is 

3. The coefficient of friction is independent of the temperature of 
the bearing. 

4. The starting effort is practically the same as the running effort. 

5. The friction in a well-designed bearing is very slightly higher 
when the bearing is lubricated than when it is dry, but in badly designed 
bearings the friction, when they are lubricated, is lower than when they 
run dry. 

6. The wear on the balls when they are not overloaded is extremely 
small and is almost negligible. 

7. There is no end thrust on ball bearings. 

8. Other things being equal, the frictional resistance with large 
balls is less than with small balls. 

9. The safe load for a given bearing diminishes as the speed of rota- 
tion of the balls is increased. 

The foregoing statement by Professor Goodman that the wear 
on ball bearings, when not overloaded, is almost negligible must 
be qualified in the light of modern practice, as ball bearings today 
are sold to last a certain number of hours. The life is inversely 
proportional to the cube of the load; thus, if the load is halved, 
the life increases eightfold. 

Common sense therefore dictates that a ball bearing be chosen 
that will last as long as the machine in which it is employed. 


Friction. In general, the friction of ball bearings is considerably less 
than that of roller bearings, but both are very much better in this respect 
than plain bearings with ordinary lubrication. 

The coefficient of friction of ball bearings is slightly less than that of plain 
bearings in a bath of oil. 

End Thrust. There is no end thrust on ball bearings, but in many roller 
bearings it is often quite serious in amount. 

Space Occupied. Most roller bearings are longer for a given load-carrying 
capacity than ordinary plain bearings. Ball bearings are, as a rule, much 
shorter and occupy much less space than even the best plain bearings. 

Shafting Mounted on Ball Bearings. For long lines of shafting, carrying 
pulleys and couplings, ball bearings are not so convenient as roller bearings. 
If a ball bearing on such a shaft fails, it is impossible to replace the ball 
races without taking down at least one length of the shafting, removing the 
couplings and pulleys, and fitting a new bearing. But with roller bearings, 
which are often used without a sleeve, there is no difficulty in replacing the 
whole bearing or any part of it without disturbing the shafting, because both 


cage and bearing liner are nearly always made in halves, a practice quite 
out of the question with ball bearings, in which extreme accuracy is required. 
With long lines of shafting, provision must be made for the expansion 
and contraction of the shaft. When plain, i.e., not grooved, bearings are 
used there is no difficulty, but with grooved bearings the outer ring must 
be mounted in a housing in which it can slide. The efficiency of power 
transmission by shafting mounted on ball bearings is higher than can be 
obtained by any other known means. 


On page 181 the various forms of friction that exist in a 
roller bearing are outlined, and it is obvious that in most roller 
bearings, owing to existing or possible end thrust, lubrication 
must be provided to reduce friction between the various rubbing 

In ball bearings there is less friction because of the absence of 
end thrust, but there is a certain amount of friction between the 
balls and the sides of the cage pockets in which they revolve. 
One form of friction that exists both in ball and in roller bearings 
has not yet been touched upon; it is due to the fact that balls, 
rollers, and races are somewhat elastic and that consequently 
instead of point and line contact we actually get metallic contact 
over a small circular and rectangular area for balls and rollers, 
respectively. The metal in front of and behind a roller, for 
example, is pushed up, as shown exaggerated in Fig. 57; the sur- 
face of the race is slightly stretched where it touches the roller; 
and when the metallic contact ceases, the surface contracts. At 
this point a certain small amount of rubbing therefore takes place 
between the roller and the race. It will be recognized that in 
front of the roller a similar small amount of rubbing takes place, 
as the surface of the race coming into contact with the roller 
becomes stretched. It will be seen that the stretching and the 
unstretching of the race in front of and behind the roller both 
tend to impede the progress of the roller and therefore create 

When the surfaces of the balls or rollers and races are very 
hard and lack elasticity, this kind of friction is less than with 
more elastic surfaces but is always very small. Lubrication of 
these points is difficult, as the pressures must be very great; 
but even if lubrication of the rolling surfaces makes them more 
slippery, it must not be overlooked that compared with dry 



surfaces we are adding a certain amount of fluid friction. It is a 
fact that the total amount of friction remains very much the same 
whether the surfaces are lubricated or not. 

As has already been mentioned, lubrication of ball and roller 
bearings is imperative to prevent rusting and to maintain the 
balls, rollers, and races in a clean 
and highly polished condition. The 
entrance of moisture and dust must also 
be avoided, so that in humid or dirty 
surroundings the bearings must have 
efficient dust guards, or they must be 
completely filled with lubricating 
grease. In the latter case a fillet of 

FIG. 57. Rolling friction. 

FIG. 58. Vertical ball bearing, 
with oil-bath lubrication. 

grease will be formed at either end which seals the bearing 
against the entrance of dust and moisture. 

Figure 58 shows the application of a ball guide bearing to a 
vertical shaft ; the housing is formed as an oil reservoir, and dur- 
ing operation centrifugal action forces the oil to rise and lubricate 
the balls. 

Figure 59 shows a vertical ball thrust bearing fitted for grease 
lubrication; with slight alteration this bearing could also be 
designed with oil lubrication without danger of oil's overflowing 
down the shaft. 

Figure 60 shows a ball thrust bearing which may be used hori- 
zontally or vertically and in which the shaft is allowed to swivel 
slightly on the surfaces indicated by the dotted circular line. 
These surfaces are ground and are submerged in oil. This 
arrangement will permit slight self -adjustment and make the 
running easier. 

Figure 61 shows an axle box with a Skefko ball bearing as used 
on a Swedish railway (Karlsbad-Munkfors Railway). The axle 



box is completely filled with grease, and it has not been found 
necessary to inspect and replenish with grease more than once 
or, twice a year. 

FIG. 59. Vertical ball thrust bearing. 

FIG. 60. Self-adjusting ball thrust bearing. 

Figure 62 shows a ball footstep bearing used for mortar mills in 
India, grinding refractory material. The dust is very hard and 
is kept out from the bearing by means of an oil seal, as shown, 
which can be removed for cleaning purposes. These bearings are 
reported to give complete satisfaction. 

It is extremely important, when handling ball or roller bearings, 
to prevent dirt, filings, etc., from getting into them; many failures 



of bearings have been caused by carelessness of this kind. When, 
for example, bearings are " cleaned" in the average motor-car 
repair shop, they are often dipped in dirty kerosene full of all 
sorts of sediment and impurities which get stirred up when the 
bearings are cleaned. 

A good cleaning agent is made from soda and boiling water 
(1 Ib. of soda to 25 Ib. of water); the bearings may be dipped 
several times to remove all grease 
and dirt, then immersed in clean 
kerosene and given a swirling 
motion, when all surfaces should 
appear bright and clean. 

Many automobile bearings 
have been ruined by wearings 
from the gears or impurities intro- 
duced when the gear case or rear- 
axle case has had its lid removed 
for inspection. Hence the design 
of oil filler as shown in Fig. 186 
(page 511) is to be recommended, 

FIG. 61. Skefko railway axle box. 

also from the point of view of the safety of the ball or roller 
bearings, now so frequently employed in gearbox or rear-axle 

FIG. 62. Oil-sealed ball footstep bearing. 

When washing motor cars with water at great pressure, it is 
quite easy for the water to enter some of the bearings (which may 
not be completely filled and sealed with grease) and cause rusting, 
with the almost inevitable result that the bearings are destroyed. 

As to whether oil or grease is to be employed, it appears to be 
preferable, wherever the surrounding air is reasonably clean and 


not too humid, to use oil. The fitting of a dust guard in the form 
of a felt packing is always advisable; the oil keeps the balls 
clean and must be an acid-free, pure mineral oil so as not to gum 
or corrode the surfaces. It should be sufficiently viscous not 
to cause excessive oil spray, but oil spray may also be caused by 
overfilling the bearings. As the friction in ball bearings is not 
influenced by the viscosity of the oil, the selection of oil may be 
entirely governed by the other conditions mentioned; of course 
when the oil is carried to the surfaces by centrifugal action it 
must not be too viscous, and at low temperatures the oil must 
have a reasonably low setting point, so as not to congeal in the 

In roller bearings, particularly those in which a certain amount 
of end thrust is created, mineral oils of heavy viscosity must be 
used for high temperatures, low speeds, or heavy loads, to 
minimize wear. Compounded oils would give better lubrication 
than mineral oils but must not be used, for the reasons mentioned 

When bearings are exposed to high room temperatures, say 
much above 140F., the oil will oxidize in time arid may produce 
a carbonaceous deposit; for such conditions, the oil must be 
changed at sufficiently frequent intervals to prevent trouble, 
whereas ordinarily the oil need not be changed more often than 
every 3 to 6 months. 

Grease is often used, and should be used, when bearings operate 
in a dusty or very humid atmosphere. The grease must fill 
the bearing cavity completely but must not be forced in so 
tightly as to impede the movements of the balls or rollers; 
high-speed bearings have been known to develop abnormal 
heat due to this cause. Replenishing with grease should pref- 
erably be done with small quantities at a time; if a big amount 
of grease is forced in by the grease gun or grease cup, trouble of the 
kind described is apt to occur. 

When the grease chamber is filled for the first time, the grease 
may be melted by gentle heat (immersion in boiling water) and 
poured into the bearing; but when high-melting-point, fibrous 
greases are used, this practice is not to be recommended. 

The grease must be as nearly neutral as possible, containing 
neither acid nor alkali, and it is essential that during manufac- 
ture it has been strained to remove all solid impurities. 


The grease must not contain any filler, as chalk or gypsum, 
nor must it contain an excessive amount of water; in good- 
quality boiled greases the water content is less than 1 per cent 
and will not cause rusting, as in grease-filled bearings the air 
has no access to the surfaces. 

Some greases are quite free from water, being simply petroleum 
jelly or mixtures of this material with mineral oil in various 
proportions. The melting points of such greases are very low; 
the melting points of boiled greases cup greases and fibrous 
greases are higher, particularly for the latter type which are 
therefore used under conditions of high room temperatures. 

The grease should be of as soft a consistency as possible, say 
No. 1 or 2 at the running temperature, so as to penetrate and 
cover all parts inside the bearings. Many automobile bearings 
have been ruined because too viscous greases have been employed, 
which cannot possibly penetrate to the points required. 

At one time many manufacturers of ball and roller bearings 
favored the use of mineral-jelly greases because of their freedom 
from moisture, but the general experience with these mixtures 
of mineral jelly and mineral oil has not been satisfactory on 
account of their deficient lubricating properties. For ball 
bearings with flat races which require hardly any lubrication, 
such greases have answered the purpose fairly well; but when 
some lubricating power is required, boiled lime greases, either 
cup or fibrous, are much to be preferred. 

For heavy-duty roller bearings such greases should be made 
from a viscous mineral oil like bearing oil 5; whereas for light- 
duty roller bearings and for all ball bearings an oil like bearing 
oil 3 is to be recommended. 

Solidified oils must never be used for ball or roller bearings, as 
they are not nearly so uniform as the boiled greases; they fre- 
quently contain a slight excess of alkali or acid in tiny pockets 
owing to the ingredients' not being so thoroughly mixed and com- 
bined as is the case in boiled greases. 

For Ball and Roller Bearings 

Bearing Oil 2.* For small and medium-size ball bearings and for small 
roller bearings with little or no end thrust. 

* For bearing oils, see p. 135. 


Bearing Oil 4. For large ball bearings and for smaller ball bearings in 
which bearing oil 2 causes excessive oil spray or leakage. 

For small or medium-size roller bearings with end thrust. 

For large roller bearings with little or no end thrust. 

Bearing Oil 6. For roller bearings heavily loaded and with end thrust 
or exposed to high temperatures. 

Cylinder Oil 2 F.S.M. (see table, page 408). For roller bearings under 
extreme conditions of pressure or temperature. 

Cup Grease 1 (made with light oil). For small ball bearings. 

Cup Grease 2 (made with light oil). For medium- and large-size ball 
bearings and for small roller bearings with little or no end thrust. 

Cup Grease 2 (made with viscous oil). For all sizes of roller bearings. 

Fiber Grease 2 (made with viscous oil) . For use in place of cup greases 1 
and 2 when the bearings are exposed to high room temperatures. 



The Michell principle described on page 174 has also been 
made use of for journal bearings, as shown in Fig. 63. The 

FIG. 63. 

FIG. 64. 

tilting pads ensure the same excellent formation of the oil film 
and low friction characteristic of Michell thrust bearings. The 
disadvantage is, however, that such a bearing allows the shaft to 
rotate in only one direction. 

The construction of the Nomy bearing overcomes this difficulty 
(see Fig. 64). The name "Nomy" indicates that these bearings 
operate with very little friction i.e., no /* (my) = nomy. 

The pads rotate with the inner ring, and every pad brings some 
oil with it from the oil reservoir in the base of the bearing housing 
(not shown). They have a spherical surface which ensures that 
they do not "edge" even if the shaft is bent or inaccurately 

If the direction of rotation is reversed, projections on the inner 
ring will move the pads so that they tilt the opposite way and 
again are in the right position for drawing in oil and supporting 
the load in the same perfect manner. 

Provisions are made for preventing the pads from moving 
sideways on the inner ring, and the special surface on the pads 
prevents them from dropping out sideways on the outer ring. As 
the inner ring and pads rotate and dip into the oil in the base of the 
bearing housing, the oil is violently thrown about and churned 
with the air inside the bearing. 




This is bound to create oil foam and oil mist the more so the 
greater the surface speed it is therefore necessary to take special 

FIQ. 65. Nomy bearing. Baffle plate arrangement. The oil that leaves the 
bearing surfaces is retained inside the fixed baffle plates K\S\ and KzSz, the oil 
reaching the narrow spherical spaces a-6 between the oil throwers OiO* 
(which revolve with the shaft) and the baffle plates is constantly forced back and 
prevented from leaking out by the centrifugal force and the adhesion to. the 
revolving oil throwers. 

precautions against aerated oil's entering the bearing surfaces and 
against oil leakage out of the bearing. 

Nomy bearings thus fulfill the following requirements which 
make them applicable for a very large range of service : 

1. They sustain the load independent of its direction, so long as the load 
is directed approximately vertically against the shaft. 

2. They operate satisfactorily notwithstanding the shaft's being bent or 
inaccurately mounted. 

3. The &xial length is so small that bending of the shaft inside the bearing 
does not affect the oil film. 

4. They support the load independent of the direction of rotation of the 

5. The coefficient of friction is exceedingly low, as the friction is entirely 

6. The wear of the bearing surfaces is theoretically nil, and the life of the 
bearing unlimited. 



Small turbines from 5 to 300 hp. operate at very high speed 
from 3,000 to 30,000 r.p.m. and are used for driving exhausters, 
exciter sets, small lighting plants, high-speed pumps, etc., both 
ashore and on board ships. 

Large stationary turbines from 300 to 50,000 hp. operate at 
lower speeds from 750 to 3,000 r.p.m. and are principally 
used to drive electric generators in electric power stations, in 
collieries, steelworks, paper mills, textile mills, etc. 

Marine steam turbines are used for the propulsion of nearly all 
warships, except submarines and some small naval craft. They 
are also used for the propulsion of steamers in mail and passenger 
service where high speed is essential. The use of a special type 
of marine steam turbine, viz. y the geared turbine, has come into 
great favor not only for warships but also for cargo boats. 

Installations have been made of from 4,000 to 70,000 hp. 
in a single ship. Marine steam turbines are frequently con- 
structed with high-pressure turbines, intermediate-pressure tur- 
bines, and low-pressure turbines, but sometimes there are only 
high-pressure and low-pressure turbines. It is usual to have two, 
three, or four propeller shafts, each shaft being driven by one or 
two turbines. On two of the shafts there are reduced-pressure 
astern turbines, which are used only for going astern. Generally, 
the low-pressure turbines are mounted on the same shafts as 
the astern turbines, close together with a common exhaust. 
Combinations may be made between reciprocating engines and 
marine steam turbines, the exhaust steam from the steam engines 
being used for operating the turbines. 

The speed of marine turbines used in the merchant service 
varies between 160 and 300 r.p.m., whereas in naval practice 
the speed may be anything up to 600 r.p.m., and on certain 
turbines in the U. S. Navy the maximum running speed goes as 
high as 900 r.p.m. 



Geared Turbines. The geared type of steam turbine has 
been used in land installations but particularly for marine services. 
Installations have been made of from 4,000 to 30,000 hp. on a 
single shaft. The turbine operates at high speed similar to the 
ordinary land steam turbine and drives by means of gearing the 
propellor shaft at low speed. The result is that the steam is 
efficiently utilized in the steam turbine, and the propeller effi- 
ciency is also high, so that the combined efficiency is considerably 
greater than where steam turbines drive the propeller shafts 

The Ljungstrom turbine is a special type of geared turbine 
operating at very high speed, say from 4,000 to 7,000 r.p.m., 
driving through gearing two electric generators. When used in 
marine service the electric current produced drives high-speed 
electric motors, say, 900 r.p.m., coupled through gearing to the 
propeller shafts (operating at, say, 90 r.p.m.). 


Parson's-type Turbines. These turbines have a great number 
of revolving and stationary disks, the steam acting on the blades 
more by its pressure than by the speed at which it impinges on 
the blades. The speed rarely exceeds 3,000 r.p.m. 

De Laval-type Turbines. These turbines have only one 
revolving disk; the steam passes through several nozzles and 
impinges on the blades with very high velocity, the action being 
similar to that of a Pelton wheel. The De Laval turbines run 
at a speed of 10,000 to 30,000 r.p.m. 

The Parsons and De Laval types of turbine represent funda- 
mentally different principles of operation, and all other types 
of turbines are adaptations or combinations of these two prin- 
ciples. The difference in design, however, affects only the 
arrangement of the revolving and stationary disks, steam dis- 
tribution to these disks, etc., and does not greatly influence the 
methods of lubrication. 

Steam. According to the steam used, turbines are classified 
as follows: 

1. High-pressure steam turbines. 

2. Exhaust steam turbines. 

3. Mixed-pressure steam turbines. 


1. High-pressure steam turbines take steam direct from the 
boilers. The steam after leaving the boilers is frequently 

2. Exhaust steam turbines principally use the steam exhausted 
from reciprocating engines, i.e., steam hammers, rolling-mill 
engines, or colliery winding engines. The pressure of this steam 
is only a few pounds per square inch. Before entering the tur- 
bine the steam passes an accumulator, which causes a steady 
flow of steam to the turbine. Exhaust steam is always very 
moist, carrying fine particles of water in suspension. 

3. Mixed-pressure Steam Turbines. Where there is not suffi- 
cient exhaust steam to operate a turbine regularly, or where the 
supply of exhaust steam varies considerably and at times becomes 
inadequate, the requisite quantities of high-pressure steam are 
automatically admitted to the turbine; hence the name " mixed- 
pressure steam turbines." 

Where exhaust steam is taken from large steam engines, it is 
important that the steam be thoroughly freed from cylinder oil 
and impurities before entering the turbine, as otherwise the tur- 
bine blades will be coated with oily deposit. The turbine blades 
can be cleaned easily by injecting at regular intervals, by means 
of a hand-operated pump, from 1 pt. to 1 qt. of kerosene per 
week. When the steam is very dirty or greasy a maximum 
amount of 1 pt. per 12 hr. should suffice. 


Owing to the high speed at which all turbines operate and to 
the fact that very little wear may cause disastrous results, the 
question of proper lubrication of the turbine bearings is of the 
greatest importance. If the oil supply fails, even for a very short 
period, or should the lubrication for other reasons become momen- 
tarily defective, the bearing in question will heat up quickly, and 
seizure will occur almost certainly before it is possible to stop the 
turbine. As a rule, turbine bearings either operate at a fairly 
normal temperature, or they are dangerously warm; for this rea- 
son every possible precaution should be taken to ensure a never 
failing supply of oil of the highest quality to each individual 
bearing, and the bearing should be carefully designed with a view 
to giving the oil every chance to distribute itself quickly over the 
entire bearing surfaces. Turbine oils must be specially manu- 


factured to withstand the destructive action of water, impurities, 
and air during continuous service. 

Lubricating Systems. Drop-feed Oilers. In the early days 
of the turbine, the bearings were fitted with sight-feed drop 
oilers, which could be regulated to give a certain number of drops 
per minute, the feed being entirely by gravity. As, however, the 
feed varied with the height of oil in the oil container, the oilers 
needed constant attention in the way of adjusting the needle 
valves controlling the feed or filling up of the oil reservoirs. 
Apart from this, the "drop-feed method " soon showed its short- 
comings when bearings were inclined to be troublesome, which 
was not infrequently the case, necessitating an increased oil feed 
and extra-close attention on the part of the attendant. 

The real cause of the small margin of safety was that the fric- 
tion was high owing to the high surface speed and, furthermore, 
that all the heat in the bearing had only one way of escape, viz., 
through radiation to the engine room from the outside of the 
bearing housings and pedestals. The bearings were always 
operating at a temperature much above that of the engine room, 
partly due to the frictional heat developed in the bearings, and 
partly due to the heat conducted through the turbine shaft. 

Ring Oiling. In modern turbine practice the drop-feed 
method has been almost entirely superseded by continuous 
force-feed oiling systems, and in the case of some few makes of 
small turbines ring-oiling bearings have been adopted for turbines 
below 200 h.p. 

A more positive system than the ordinary ring-oiling arrange- 
ment is illustrated in Fig, 66, the oil ring (1) having a "U" 
section, and the oil being continuously diverted into the bearing 
by the stationary oil scoop (2). If the oil well contains a fair 
quantity of oil, the heat can be readily conducted to the bearing 
pedestal and radiate into the engine room without the bearings' 
getting uncomfortably warm. Water cooling of the oil has been 
resorted to in some cases with very good results: (a) in the shape 
of a cooling coil in each bearing pedestal; (6) by casting the two 
bearing halves with cavities for water circulation; or (c) by having 
a central oil cooler and an oil pump forcing the oil through the 
cooler and thence into the various bearing oil wells. The oil 
overflows from each bearing back into the tank from which the 
oil pump draws its supply and circulates the oil afresh. Using 



ordinary ring-oiling bearings without water cooling is, of course, 
cheaper than a forced-feed circulation system but does not give 
the same margin of safety. Care should be taken that the oil is 
changed at intervals of, say, 3 months. If the oil is of good 
quality, it can be used over again after proper separation from 
water, dirt, and other impurities in a steam-heated settling tank, 
followed by filtration through an efficient steam-heated filter. 

Force-feed Circulation. This system is in general use for 
practically all turbines above 200 hp. Only in very rare cases 

FIG. 66. 

has the oil been forced into the bearings at the points of greatest 
pressure, lifting the shaft of the rotor and thereby keeping it 
"floating." In order to accomplish that, an oil pressure some- 
what higher than the maximum bearing pressure per square inch 
is required. If several bearings are fed from the same oil-dis- 
tributing pipe, they must all have approximately the same load 
per square inch, as otherwise the bearing with lower bearing 
pressure would rob the other bearings of a portion of their share of 
the oil supply, the oil naturally taking the way of least resistance. 
The term "force feed" therefore generally means that the oil 
is kept in circulation by means of a pump at a pressure usually 
much below the bearing pressures. The oil is introduced at the 
top, or "on," side of the bearings and wedging itself in between 
the revolving shaft and the bearing surfaces produces a complete 
oil film on which the whole weight of the revolving part "floats." 


If a continuous flow of oil through the bearings is provided, the 
oil carries away not only the greater portion of the f rictional heat 
but also the heat conducted through the shaft from the highly 
heated parts of the turbine. The combined loss from friction 
and heat transmission into the bearings is estimated at about 
J per cent of the rated horsepower of the turbine. 

The Oil Cooler. It therefore becomes necessary to cool the 
return oil from the bearings, and it cannot be too much empha- 
sized that an efficient, well-designed oil cooler of ample capacity 
is one of the best investments that can be made in a turbine 
plant and is an excellent insurance against lubrication troubles. 
There is a variety of designs of oil coolers. In the early days they 
were often " built in" in the bedplate. This practice seems now 
to be practically abandoned; because of the proximity of cold 
water and hot oil, extra stresses are set up in the turbine bed- 
plate, owing to the unequal expansion of the various parts, and a 
cracked bedplate has occasionally been the result. 

Another reason for building the oil cooler separate from the tur- 
bine is the vibration which tends to disturb joints, etc. One 
curious result of heavy vibration was the wearing through of a 
cooling coil rubbing against the bottom of the oil cooler; it was 
finally perforated, and the water leaking into the oil caused con- 
siderable trouble. When a new coil was fitted it was raised above 
the bottom and had small "feet" clamped on to it at intervals. 
This successfully overcame the trouble. 

Oil coolers should be designed with a view to facilitating 
inspection and cleaning of the tubes, internally as well as exter- 
nally, and the water spaces. The oil cooler should always have 
doors for inspection, large enough so that the tubes can be cleaned 
on the outside. The tubes should be solid drawn, seamless, with 
no unnecessary connections, which might cause leakage; fre- 
quently they are so arranged that they can be withdrawn as a 
whole for inspection and cleaning. 

In the earlier type of coolers the tubes were usually of the 
U type, but most modern coolers have straight tubes, which 
are easier to clean. The flow of oil and water through the cooler 
should always be in opposite directions, so that the oil in passing 
through meets colder and colder water; in this way the best 
cooling effect is obtained. In most coolers the oil is inside the 


It is highly desirable that the pressure of the oil in passing 
through the cooler should at all points be higher than the water 
pressure, so that should any leakage occur it will be of oil into 
water; otherwise, it will mean water leaking into the oil, which 
must be avoided for reasons given later on. 

The capacity that hot oil/ in particular, possesses of percolating 
through the most minute pores or leaks is remarkable, and leakage 
may occur under running conditions, even if the cooler has been 
tested cold and found perfectly tight under great pressure. When 
testing an oil cooler for leakage it should, therefore, always be 
tested "hot." 

The cooling coils sometimes get badly corroded when acid 
water is used, and corrosion nearly always attacks certain spots 
in the tubes, particularly if the latter are made of brass. It 
looks as if local galvanic currents may often have something 
to do with heavy corrosion, caused by inequalities in the com- 
position of the tube metal and due to the presence of small grains 
of different metals close together copper, zinc, etc. To pre- 
vent corrosion in oil coolers employing sea water, an iron rod 
fixed from end to end of the cooler has proved effective; the 
rod is often eaten away by galvanic action in a single voyage. 

The cooling water should preferably be circulated through the 
cooler by means of a circulating pump and at a low pressure, 
which falls to nil when the turbine stops running. The efficiency 
of the oil cooler is greatly affected by dirty cooling water; cases 
have been known where greasy, muddy river water caused by 
dirty discharges from works higher up the river or due to heavy 
rainfall used as cooling water has deposited sufficient slime and 
dirt to increase the turbine oil temperature at the rate of 1F. 
per day. 

The oil cooler has its best place in the circulation system 
after the oil pump, not before, 1 as in the latter case the oil is 
sucked through the cooler, and any leakage would be of water 
into oil. 

Thermometer pockets should be fitted in the inlet and outlet 
oil pipes, also in the water inlet and outlet to the oil cooler, as by 
temperature records taken, say, every hour it will at once be 
discovered if there is anything wrong with the cooler or with the 
oil in circulation. The water, if not clean, may have thrown down 
muddy deposits on the tubes, or the tubes may have been coated 


on the "oil" side with deposits from the oil system. In any case, 
the temperature records will at once indicate whether trouble is 
approaching, and a close investigation in good time will locate 
the cause and point out the remedy. 

Shutting off the cooling-water supply is the last operation when 
stopping a turbine, but the oiling should be continued until the 
turbine has come to a standstill. 

If an oil cooler is found to be too small in capacity, it is not 
of much use to increase the flow of cooling water through the 
cooler; it will, of course, make some difference, but if the cooling 
water is taken from the coldest available supply, and if the oil 
does not get cooled sufficiently, the only remedy is to increase 
the capacity of the cooler by adding more "surface." 

In some installations where the oil is inside the tubes an im- 
provement has been made by fitting twisted strips of the same 
material as the tubes inside the tubes in order to disturb the 
oil as much as possible; it is obvious that as long as the flow of 
oil is only 1.0 to 2.0 ft. per second, which represents normal 
practice, the oil ordinarily shoots through the tubes without being 
"broken up," and a layer of cold oil on the inside of the tubes 
makes the cooling of the oil in the center rather inefficient. The 
value of inserting the twisted strips retarders will be easily 

The Oil Pump. The development has been in the direction of 
rotary, toothed-wheel pumps driven by worm or skew gearing 
off the main turbine shafts. The toothed-wheel pump is more 
positive in action than the valveless "sliding- vane" type of 
pump; also, there is less chance of the toothed-wheel pump's 
being accidentally choked with rusty scale, dirt, etc., as the oil 
has a more effective washing action in passing through the pump. 
On the other hand, the toothed-wheel pump has the disadvantage 
that the oil is "churned" more vigorously and may consequently 
suffer more when water happens to be present. 

The oil strainer consists of copper or brass gauze, supported by 
a perforated cylinder, which it covers. This cylinder should be 
of the same metal as the gauze, as otherwise galvanic action comes 
into force and destroys the strainer by pitting and corrosion. 
The oil strainer should be situated in a position well clear of the 
bottom of the oil tank, say 4 to 6 in., to allow the water which 
almost invariably leaks into the oil to separate out, so that it can 


be drained away through a drain or sludge cock of ample dimen- 
sions not less than 1^-in. bore. The need for such a big bore is 
on account of the sludge, which may be formed in the oil, and 
which will not drain out through a small opening. If the 
strainer is placed close to the bottom of the tank, water is sucked 
into the pump first of all and is not given a chance to separate 
out from the oil. Care should be taken to have sufficient oil 
above the top of the strainer so that no air can be drawn in with 
the oil, as aeration of the hot oil has an oxidizing effect and 
may cause decomposition of the oil, if the temperature is high. 

In large installations the oil pumps are nearly always operated 
separately from the turbine, either electrically or by steam; the 
pumps are started up before the turbine and kept in operation 
until the turbine has come to a standstill. In smaller installa- 
tions, where the oil pump is an integral part of the turbine, the 
pump will not supply a sufficient quantity of oil until a certain 
speed has been reached; it therefore becomes necessary to have 
an auxiliary oil supply which works independently of the turbine- 
driven oil pump. This auxiliary supply is usually a hand pump, 
with which the bearings can be flushed before and during the 
starting up of the turbine ; in larger installations a hand-operated 
pump becomes inadequate, and the auxiliary oil pump is driven 
by an electric motor or by steam. It cannot be too strongly 
emphasized that the bearings should be continuously flushed 
until the speed of the turbine is about 20 to 25 per cent of the 
normal; this should also be done when in stopping the turbine the 
speed has fallen to the speed just mentioned. By watching 
the pressure gauge attached to the main circulating system, 
the attendant can always be guided as to the time when it will be 
safe to discontinue the auxiliary oil supply. 

The oil pump should be designed to give a supply of oil at the 
pressure required, equalling 0.05 to 0.15 gal. per min. per square 
inch of total projected bearing surface. 

The strainer on the suction side of the oil pump should have 
an area in square inches equalling from four to six times the 
number of gallons of oil circulated per minute. 

The quantity of oil present in the circulation system should be 
from 0.15 gal. per kilowatt for smaller turbines to 0.10 gal. per 
kilowatt for the largest units, but the minimum amount of oil 
in any turbine should preferably be 120 gaL 


Oil Pressure and Circulation Systems* On leaving the oil 
cooler, the oil generally goes to the main oil-distributing pipe, 
which runs along the turbine bed and from which branch pipes 
lead it into the various bearings. It is forced into the main oil 
pipe under a certain pressure which is regulated by means of a 
spring-loaded relief valve; this valve can be regulated to give 
any desired pressure within certain limits, and the surplus oil is 
allowed to discharge back into the suction oil tank. 

Another way of maintaining a certain oil pressure is to force 
the oil up into an elevated tank, from which it is led through a 
main pipe down to the turbine and then distributed in the ordinary 
way. This system has the advantage that, should the oil pump 
fail for some reason or other, the top tank will continue the supply 
for a sufficient length of time to allow the turbine to be shut down 
before any damage is done. The top tank should be fitted with 
an overflow pipe to carry surplus oil down into the return oil tank. 
It should also have a drain or sludge cock of at least IJ^-in. bore 
and a drain pipe. 

The return-oil tank must be of sufficient capacity to take the 
whole of the oil in the system, in case during a standstill the whole 
of the oil in the top tank should be allowed to run down into the 
bottom tank. 

Sight feeds are sometimes fitted in the inlet branch pipes 
to the bearings, their position being between the bearings and 
regulating cocks fitted in the inlet pipes. It is, however, difficult 
to keep them clean. If the oil drops through the sight feeds, it 
has a tendency to take the air away, and the sight feeds fill up 
with oil. The same unsatisfactory results are generally experi- 
enced where the sight feeds are filled with water and the stream 
of oil is made to rise through the water; the oil carries the water 
away, and the glasses fill up with oil. 

It is, of course, very desirable to have efficient sight feeds, but 
it is preferable to fit them in the return-oil pipes from each 
bearing. Care should be taken that the outlets have ample 
openings to allow of the oil's running through freely; otherwise it 
cannot escape from the bearings quickly enough and overflows 
through the bearing ends. Some makers just fit onto each bearing 
a small test cock which if opened shows whether the inlet pipe is 
supplied with pressure oil or otherwise, but, of course, this does 
not give any idea as to how much oil goes through each bearing. 


In plants with a top oil tank, distributing the oil by gravity, the 
oil pressure is a fixed figure; but where the oil is distributed 
direct from the oil pump, the maximum oil pressure that can be 
obtained depends upon the capacity of the pump and the resist- 
ance offered to the oil in its passage through the oil pipes and 
the bearings. The warmer the oil or the thinner the oil in use 
the lower will be the maximum oil pressure obtainable, as the 
thinner oil passes more easily through the bearings and leaks 
more freely in the case of a rotary oil pump; i.e., the pump dis- 
charges less oil per revolution. 

A lowering of the oil pressure of a dangerous nature may take 
place when unsuitable oil gets very thick and sludgy, owing to 
emulsification with water, particularly if the pump strainers are 
covered with sludge. Under these conditions a vacuum is 
formed in the pump; it " slips" and does not operate with its 
full capacity; hence the oil pressure falls, sometimes with only 
short warning, and may cause disastrous results. When a 
turbine starts up cold, the oil pressure is usually high, even if 
the relief valve is fully open; as the oil warms up, the pressure 
falls but should not fall more than can be met by partly closing 
the relief valve when the running temperature has been reached, 
thus leaving a margin over and above the minimum pressure 
required to operate the governor gear. 

The cooling-water service should not be put on until the oil 
in circulation has warmed up to within 20F. of its normal 

As regards the oil pressure required for distributing the oil 
with certainty to the bearings, a few pounds per square inch will 
be found adequate to give a satisfactory supply, say from 5 to 
15 Ib. per square inch. 

In many modern designs of turbines the oil is made use of in 
several other ways, the principal one being in connection with the 
operation of the governor gear. It is beyond the scope of this 
book to go into the various designs of oil-worked governor gears, 
but the author would like to emphasize the necessity of using 
good-quality oil and keeping it in first-class condition; otherwise, 
the result may easily be sluggish and unsatisfactory govern- 
ing, as some of the clearances are exceedingly fine. One point 
worth mentioning is that, where a vertical spindle operates the 
steam-throttle valve below, the oil-worked piston being above 



(see Fig. 67), the stuffing box of the oil cylinder invariably leaks 
slightly. If this oil is allowed to trickle down on top of the throt- 
tle-valve cover, it will smoke and "bake on" in the form of a 
carbon deposit, particularly when superheated steam is used; this 
is rather objectionable and may easily be prevented by fixing a 
cup on the spindle and a drain to carry oil away outside the 
throttle valve, as shown. 

1 Preiiaie Oi/ Inlet 

2 PreMure Oil Outlet 

3 Bpringloaded Pitton 

4 Throttle Valre Spindle 

5 Oil Drain 

Fio. 67. Turbine throttle valve. 

The principle of operation of oil-worked governors embodies 
a pilot valve, which is moved by the governor, and which when 
moved allows pressure oil to flow into an oil-relay cylinder, 
thereby causing a spring-loaded piston to rise or fall in this 
cylinder, according to whether the oil is introduced above or 
below the piston or, if acting only on the underside, according to 
whether the oil is admitted or not. The piston, moving with 
great force, acts directly on the main steam valve. When the oil 
supply to the governor is taken from the main oil-circulation 
system, a failure of this oil supply will cause the relay piston to 
descend and shut off the steam supply operating the turbine. 


The turbine cannot start until a sufficient oil pressure has been 
obtained in the oil-supply system, and consequently any damage 
to the bearings due to insufficient pump pressure is thus obviated. 

The oil pressure required by the governor gear is high from 
25 to 60 Ib. per square inch in accordance with the requirements 
of the various designs. 

Several makers fit two oil pumps one delivering small 
quantities of oil under great pressure for the governor gear, the 
other delivering large quantities at low pressure for lubricating 
the bearings. 

Oil Pipes. The distributing-oil pipes should be of ample pro- 
portions, with as few bends as possible. The branch pipes lead- 
ing to the bearings should not join the main oil pipe at right 
angles but preferably at an angle not more than 30 deg., with a 
view to decreasing the loss in oil pressure due to fluid friction and 

As regards the return-oil pipes from the different bearings, 
they should be of ample proportions, so that the maximum 
quantity of oil from each bearing may return comfortably; 
otherwise, the oil may overflow through the bearing ends and 
cause unsightly waste of oil, with a possibility, in the case of 
turbogenerators, of the oil's being drawn over into the generator 
and spoiling the insulation. The branch pipes should meet the 
main return pipe at an angle of not more than 30 deg. ; and in case 
of sight feeds in the branch pipes, these should be designed so 
that air does not get churned with the oil, causing aeration. At 
no place must the flow of oil be broken up or violently disturbed. 

During late years most turbine makers have made the oil 
pipes of steel or wrought iron instead of copper, which originally 
was used exclusively. This has been done largely in view of 
lower first costs, but it is very questionable whether this step is 
an altogether wise one. The oil is nearly always charged with 
finely divided globules of air and water, and through the con- 
tinued use it always becomes slightly acid. These features 
combined cause corrosion of the iron or steel pipes in a much 
higher degree than when the pipes are made of copper in fact, 
copper is hardly affected (see page 226 and Example 6, page 232). 

Bearings. The load on the main bearings of a turbine is 
due mainly to the weights of the rotor, shaft, and generator, if 
any. The pressure is therefore the same, whether the turbine is 


under load or otherwise, and is never relieved, as is the case with 
the principal bearings of reciprocating engines. It is conse- 
quently of the very greatest importance to design the bearings 
with a view to quick distribution of the oil, particularly in the 
case of marine turbines, where greater pressures per square inch 
are carried in connection with lower surface speed. A high 
surface speed draws the oil in between shaft and bearing and 
makes it possible, and desirable, to use free-flowing oils. On the 
other hand, the lower surface speed and higher bearing pressures 
met with in marine practice necessitate the use of heavier bodied 
oils and may even make careful oil grooving desirable. 

With high surface speed, oil grooves should be dispensed 
with altogether, being distinctly detrimental, as they reduce 
the area of the bearing surface. 

In some cases turbine bearings have been designed as oil- 
cooled bearings, the oil before entering the frictional surfaces 
first passing round the outer surfaces of the bearing shells. 
The result is that the bearings are kept at a uniform temperature 
throughout and that the oil removes a little more heat than it 
would have done had it been passed direct into the frictional 

Should a bearing give trouble, it generally gives no warning; 
the oil evaporates, and white fumes ooze out from one or both 
ends. The turbine should be stopped at once, as the white 
metal with all certainty has started to run and will want renewing 
before the turbine can be put under load again. Grit or dirt 
may have been the cause. Failure of the oil supply, if due to 
the oil pump's pumping an insufficient amount of oil, will show 
up in decreased oil pressure and should be noticed by the attend- 
ant. Choking up of one of the oil-distributing pipes to a particu- 
lar bearing might be noticed in time, if the bearings are fitted with 
sight-feed attachments. 

Whenever a bearing cap has been adjusted, the turbine should 
not be put on full speed until one is fully assured that the bearing 
does not pinch the shaft. 

The amount of oil circulated per minute varies according to 
the oil pressure required and to the size of the bearings. Current 
practice is to circulate the oil at the rate of 0.05 to 0.15 gal. per 
min. per square inch of total bearing surface, as mentioned under 
the heading "Oil Pump." In the case of slow-speed marine tur- 


bines, a supply of 0.02 gal. per min. per square inch will be found 
adequate. This lower rate of feed emphasizes, however, the 
desirability in the case of marine turbines of having oil sight feeds 
in the return pipes from each individual bearing to make sure that 
each bearing gets its proper share. 

Turbine Thrust Bearings. Where greater or smaller end 
pressure has to be taken up, owing either to the design of the tur- 
bine itself or to propeller thrust, the thrust bearing becomes an 
important feature of the turbine. 

Thrust blocks for marine turbines are usually of cast iron, with 
a steel bush for holding the thrust rings. The top portion of the 
thrust block generally takes the steam thrust, and the lower por- 
tion the propeller thrust. The block is fitted on a sole plate of 
its own and can be moved in a fore-and-aft direction; also, the 
upper portion can be moved relatively to the lower portion in 
order to adjust the clearances of fore-and-aft play, which may be 
made about 0.01 in. The thrust rings may be made of gun 
metal with white-metal facings on the rubbing surface. 

It is evident that when the thrust block is supplied with oil 
under pressure from the outer edges of the thrust rings, the oil 
has to go against the action of the centrifugal force, and when it is 
between the rubbing surfaces the tendency is to squeeze it out all 
the time; whereas in the main bearings of the turbine the revolv- 
ing shaft draws the oil in between the rubbing surfaces, feeding 
it toward the place where it is needed. An increased oil pressure 
does not help the oil in the case of a thrust block; the oil has only 
its natural clinging property oiliness to depend upon for 
getting to the place where it is required. 

Thrust blocks in steam-engine-propelled ships are lubricated by 
means of oils heavily compounded with vegetable oils. The 
reason is that such oils, properly manufactured, have very great 
clinging properties, so that they are able to get in between 
the rubbing surfaces better and more easily than pure mineral 

In forced-lubricated thrust blocks in connection with marine 
turbines the oil is taken from the main circulation system, as it 
would be cumbersome to make a separate oiling system for the 
thrust blocks. But oils used for forced lubrication must be pure 
mineral in character, and, in view of what is said above, it is 
obvious that a heavy-bodied oil will be needed for the thrust 


blocks, as light-bodied pure mineral oils would cause the thrust 
bearings to run hot. 

Another condition in connection with marine turbines that 
calls for more viscous oil than similar-size land turbines is the 
vibration, which is set up partly by the turbines themselves and 
partly by the reaction of the water on the propellers. Heavier 
vibration calls for better cushioning in the bearings, and this can 
be given only by employing a more viscous oil. 

Thrust bearings of the ordinary type carry a maximum bearing 
pressure of 15 to 20 Ib. per square inch in the case of land turbines 
and 30 to 50 Ib. in the case of marine turbines. 

Attempts have been made to introduce actual forced lubrica- 
tion conditions in the thrust bearings, by making the oil pass 
through the hollow shaft and thence forcing it out between the 
revolving thrust rings and the stationary thrust block. Such a 
system has been designed by Ferranti and is said to have given 
excellent results, making it possible to carry great pressures 
without any fear of the bearing's seizing. 

An ingenious method of getting over the difficulty with the 
thrust bearing has been designed by Franco Tosi. He balances 
the difference between the propeller thrust and the steam thrust 
by means of oil pressure exerted on the two sides of a piston which 
revolves with the shaft and is fitted with a labyrinth packing. 
Oil under pressure is constantly being forced into the chambers on 
both sides of the piston and can escape only between the collars 
of the thrust blocks at either side. If the thrust is from right to 
left, the clearances on the left-hand side are diminished, so that it 
is easier for the oil to escape between the right-hand thrust 
collars; consequently, the oil pressure becomes lower in the right- 
hand chamber, and the difference in oil pressure forces the piston 
to the right, or vice versa, thus automatically balancing the axial 
thrust and preventing metallic contact between the rings and 
the blocks. At high speed, fluid friction developed between the 
piston and its casing, etc., would be very considerable; but as 
marine turbines are slow speed, this loss is only small. 

With Parsons steam turbines the axial thrust on the rotor is 
more or less balanced by the propeller thrust, and the thrust 
bearing embodied in the turbine itself gives no great difficulty; 
but with geared turbines, with the reintroduction of a main thrust 
block on the propeller shaft, the multicollar marine type of 
thrust bearing has failed to give satisfaction, 


The even turning moment of the turbine transmitted through 
the gearing never pulsates or fluctuates, thus not giving the thrust 
collars that relief which in the case of a steam engine in 
some measure helps the oil to creep in between the rubbing 

For geared turbines, the thrust problem has been solved by the 
Michell single-collar thrust bearing, described on page 175, which 
will carry a bearing pressure of 400 to 500 Ib. per square inch 
with the greatest ease and with a surface speed ranging from 
60 to 100 ft. per second. 

The Hon. Sir C. A. Parsons has designed a similar type of 
bearing, but with centrally pivoted segmental blocks, allowing the 
turbine shaft to revolve in either direction. The frictional losses 
in these types of bearings are considerably less than in the ordi- 
nary plain type of thrust bearing; the coefficient of friction may 
be taken as 0.002 as against 0.03 to 0.04 for ordinary thrust 

Wear. As turbine bearings are virtually flooded with oil, 
it is probable that the shaft never comes into actual rubbing 
contact with the bearings except at the moment of starting. 
When the turbine is standing, the oil film is pressed out, and actual 
contact between journal and bearing probably takes place; but 
as soon as the turbine starts running, the first few revolutions 
will build up the oil film, which, if the oil is satisfactory, will 
support the shaft; i.e., it "floats" on the oil film. 

Turbine bearings, i.e., the vast majority, practically never 
wear. It sometimes happens that what may appear to be wear 
takes place for a certain length of time, after which it ceases; 
this is, in reality, due to compression of the white metal, which 
has been rather soft. 

After many years' working, the toolmarks should still be 
visible if the turbine has had proper care and attention. 

Temperature Records. When a turbine starts from cold, the 
oil will gradually rise in temperature, rapidly at first, slowly later 
on; and if the conditions remain fairly uniform uniform load, 
uniform temperature of cooling water and engine room the 
maximum temperature will be reached after a certain number of 
hours, varying from 4 in the case of small turbines to 8 hr. or 
even longer in the case of large units. This final temperature is 
not much affected by changes in the engine-room temperature or 


The temperature of the cooling water, however, and the state 
of cleanness of the oil cooler have a marked influence on the oil 
temperature, and naturally so, because it is in the oil cooler 
that the bulk of the heat is removed from the oil, a minor portion 
only being radiated into the engine room from the bearings, 
pedestals, oil pipes, oil tanks, etc. 

The temperature of the oil in the main return pipe ranges from 
100 to 140F. seldom below 100 and not often above 140F. 
But bearings have run without trouble for long periods at tem- 
peratures as high as 160F. However, it is desirable that the oil 
temperature should be about 120 to 130F. 

In the case of marine turbines, the oil temperature rises in the 
tropics as compared with conditions in temperate climates, 
owing to the higher temperature of the cooling water, being 70 
to 85F. as compared with 50 to 70F. for temperate climates. 
The oil must, of necessity, rise in temperature in order that the 
difference in temperature between itself and the cooling water 
may enable the water to take away the heat from the oil. 

It would be useless to quote actual temperature records from 
turbines in operation, as they vary exceedingly; e.g., in some 
turbines the fall in temperature between the return oil and 
the oil leaving the cooler is as low as 4F., owing to the fact that 
the amount of oil delivered by the oil pump to the bearings is high 
per square inch of bearing surface and to the fact that the bearing 
is water cooled. 

In a good many turbines the difference in temperature between 
the outgoing and returning oil is from 15 to 20F., and in some 
extreme cases, where the bearings have not been water cooled and 
the oil deliveiy per square inch of bearing surface has been low, 
this difference has been as high as 50F. The temperature 
rise of the oil going through the worm-wheel casing (the worm- 
wheel shaft operating the oil pump, governor, and sometimes the 
circulating pump for the condensing plant) or the thrust bearing 
is usually considerably higher than the temperature rise of the 
oil going through the other bearings. 

For each particular turbine the temperatures of the oil and 
water inlet and outlet to the cooler and of the oil inlet and 
outlet to the various bearings or, as an alternative, the bear- 
ing temperatures at both ends of each bearing indicate whether 
normal conditions of lubrication and cooling prevail. 


Thermometer pockets filled with oil should be fitted in the 
positions mentioned above. In case of the return-oil tempera- 
tures taken in the bearing outlets, care should be observed that 
the flow of oil shall wash over the thermometer bulb or the pocket. 
Occasionally, the thermometers should be compared with a 
standard thermometer, say once a year. A temperature log 
should be kept in the engine room, taking the temperatures 
every hour. 

The turbine attendants will very soon get to know by heart 
the normal running temperatures at the various points, and they 
will learn to interpret the correct causes of any deviations from 
the normal temperatures or, at any rate, to look in the right direc- 
tion for the cause of irregularities, indicated by abnormal 

The Turbine Glands. The most frequent cause of water's 
getting mixed with the oil in circulation is leakage of steam 
past the glands, the steam condensing on the shaft and bear- 
ings, gradually working its way into the main bearings and 
mixing with the oil. It will, therefore, be useful to look for a 
moment on the various designs of glands. 

There are three types: 

1. The labyrinth packing gland. 

2. The carbon packing gland. 

3. The water-sealed gland. 

The function of the gland is either to keep high-pressure steam 
from leaking outward or, in the case of the " vacuum end" 
of the turbine, to prevent air from being drawn in, which would 
adversely affect the vacuum produced by the condensing plant. 

1. The labyrinth packing (Fig. 68) consists of a series of rings 
on the shaft which alternate with stationary rings in the sur- 
rounding casing; there is only a very slight clearance between 
the shaft and the stationary rings. 

The steam in the case of a high-pressure gland must pass 
the rings in a zigzag way, so that only a slight amount of steam 
escapes at the vent (1), which may be connected with an inter- 
mediary stage of the turbine or simply allows the steam to escape 
into the open. Any steam or water leaking outside the gland is 
deflected by suitable throwers fixed on the shaft in order to 
prevent the water as much as possible from getting into the adja- 
cent bearing. 



In the case of a low-pressure gland, steam at a reduced pres- 
sure either from an intermediary stage or high-pressure steam 
throttled down to the required pressure is introduced at (1) 
and leaks inward into the low-pressure turbine casing. 

FIG. 68. Labyrinth packing. 

The pressure should be sufficient to prevent air from leaking 
in; i.e., sufficient pressure should be applied so that a puff of 
steam is just visible oozing out of the gland. Excessive pressure 
should be avoided, as it means not only waste of steam but also 
excessive condensation of steam on the shaft with a certainty 
of some of this getting into the bearings. Avoiding excessive 
condensation is particularly difficult in the case of the low-pres- 
sure glands of exhaust-steam turbines with labyrinth packing 
glands. Owing to the variation in steam pressure, it becomes 



necessary for the attendant constantly to readjust the gland 
pressure; otherwise, either air will occasionally leak inward, or 
excessive leakage of steam will take place outward. Occasionally 
the glands are water cooled, as the steam then condenses on its 
way through the gland, and consequently the thrower outside 
the gland has to deal only with water, which can be much more 
effectively thrown away from the shaft than steam. 

FIG. 69. Carbon packing. 

2. The carbon packing (Fig. 69) consists of a series of carbon 
rings, each made up of several sections and held in their places 
around the shaft by means of springs. The carbon rings should 
preferably not bear right against the shaft but on a special 
sleeve fixed on the turbine rotor and slightly larger in bore than 
the turbine shaft, so that, should heavy wear take place, the shaft 
will remain unhurt, and only the sleeve or the packing itself will 
get worn. No lubrication is needed of the carbon rings, but care 



should be taken that they be a loose fit on the cold shaft, as 
carbon contracts when heated. If grit and dirt get in, cutting 
and wear may occur owing to the absence of lubricant, and 
then leakage through the packing will take place. Sometimes 
the carbon packing glands are surrounded with a water jacket, 
which causes a certain amount of steam to condense in the pack- 
ing; this helps to seal the gland and "lubricate" the carbons. 
The vent (1) serves the same purpose as in 
Fig. 68. 

3. The water-sealed gland (see Fig. 70) 
consists of a revolving wheel (1) formed with 
vanes on both sides and acting like a centrif- 
ugal pump. The water admitted at (2) (or 
sometimes at the circumference at (3) under a 
few pounds pressure) is thrown by centrifugal 
action to the outer edge of (1) and thus estab- 
lishes a perfect seal, it being impossible for 
steam to escape round the outer edge, the 
clearance being about 0.01 to 0.02 in. The 
water should be clean and preferably soft, as 
otherwise dirt or scale will be deposited in the 
gland and may even get inside and coat some 
of the turbine blades. The water supply 
should be kept as low as possible by regulating 
the quantity admitted. The second disk (4) 
revolving in the groove (5) acts as another seal 
in series with (1), but the chief object in fitting 
it is to prevent water from escaping from the 
gland past the groove (5). The water coming 
into this groove will be drained back into the main gland through 
drain holes (6) indicated at the bottom. 

During the time the turbine is being warmed up prior to start- 
ing, and where water-sealed glands are employed, the vacuum 
cannot be created until the turbine has attained a certain speed, 
as the glands do not provide a perfect seal until the centrifugal 
force is sufficient to prevent the air from going straight into the 
turbine. In case of high-pressure water-sealed glands, it is 
frequently desirable or necessary to water cool the gland casing, 
as it otherwise becomes so warm that the water evaporates too 
readily and gets into the turbine in the form of steam; and, also, 

FIG. 70 . Water- 
sealed gland. 


should the water not be very soft, a certain amount of scale will 
be deposited, which is objectionable. It is only at low speeds 
starting and stopping that the high-pressure water-sealed glands 
allow steam to escape, thus making it possible for condensed 
steam to enter the bearing nearest the gland and mix with the 
oil in circulation. 


The idea of the geared turbine of large horsepower is to run 
the turbine at high speed, transmitting the power through double 
helical gearing to a low-speed propeller shaft or generator 
shaft thereby getting a very high over-all efficiency. The 
gears, when not very accurately made, are noisy and inclined to 
wear, but the latest developments seem to be overcoming all 
obstacles in this direction. 

If the gears are perfect, and as long as they remain so, the oil 
used in the turbine can also be used for them, being constantly 
supplied in streams at the points of contact between the teeth. 
But if the gears are inclined to be noisy, a heavier oil will be 
preferable in order to deaden the noise. Such a heavy oil will 
not be satisfactory in the turbine system, as it will separate 
only slowly from water and dirt and cause high temperatures 
all around. 

If one oil system only is used for turbine bearings and gearing, 
and if the oil gets mixed with water from the glands or the 
cooler the oil will suffer in the turbine system to some extent; 
but when this same oil, mixed with minute particles of water 
and dirt, gets through the gearing exposed to many times the 
ordinary bearing pressure, it is sure to suffer very quickly indeed, 
and the result will be wear of the gearing. For these reasons, 
the author strongly recommends that the oiling system for the 
gearing should be made distinct and separate from the oiling 
system supplying the turbine bearings, quite apart from the 
question of whether the same oil or two different oils are used in 
the two systems. With separate oiling systems, the oil for the 
gears will remain dry and pure for a much longer time and will 
thus have a much better chance of keeping the teeth of the gears 
in good condition and preventing wear. 

Treatment of the Oil. Before starting a new turbine, it 
should be carefully cleaned all through the oil tanks, oil pipes. 


etc., in order to remove as much grit and dirt, molder's sand, 
rusty scale, cotton waste, etc., as possible. Cotton waste must 
never be used for cleaning purposes, as it leaves behind small 
fluffy pieces, which will tend to clog up the oil pipes and particu- 
larly the fine clearance spaces in the oil-worked governor. 

Mutton cloths or sponges should be used for cleaning, and it is 
preferable to use a cleaning oil light petroleum distillate with a 
higher flash point than paraffin rather than paraffin, as some of 
the oil remains and mixes with the lubricating oil. Paraffin will 
start to evaporate when the turbine starts running and may 
cause an explosion. The air should be driven out of the oil 
piping by means of the auxiliary oil pump, and when the pump 
is being filled with oil it should be put through the sieve and not 
direct into the tank, although the latter may be the quicker 

After a new turbine has been run a month, during which time 
frequent examinations of the oil strainer will prove of interest, 
the whole charge of oil should be removed, and the oil tank and 
oil pipes, as well as the bearings, again thoroughly cleaned. The 
oil taken out, in which will be found impurities of many kinds, 
such as cotton waste, rust, sand, dirt, little pieces of iron, copper, 
red lead, and packing material, should be treated in a steam- 
heated separating tank and afterward in a good steam-heated 
filter or a centrifugal oil purifier. It can then, if it was originally 
of good quality, be used as " make-up " in the circulation system, 
which in the meantime has been filled with a fresh charge of oil. 
This first change of oil may seem an unnecessary precaution to 
take, but it is the author's strong recommendation, based on 
long experience, that it should always be made and that it pays 
in the long run. 

It is during the early life of a turbine that it needs the greatest 
amount of care and attention; later on, troubles are or ought to be 
rare if the oil is well looked after, frequently filtered, and the 
strainers kept clean. As regards the inside of the turbine oil 
chambers, etc., the surfaces have by some makers been painted; 
this has sometimes been done in order to save the labor of 
cleaning and scraping the surfaces. In nine out of ten cases 
the paint itself has been by no means oilproof , and the result 
has been that the warm oil quickly dissolved it, causing long 
protracted troubles with the oil breaking down and carrying 



sticky brownish-black deposits everywhere throughout the oil 
system. The writer recommends leaving the tanks, etc., 
unpainted but that the surfaces should be very carefully scraped 
and cleaned. Sandblasting appears to be too " searching," 
small grains of sand being embedded in the cast-iron surfaces 
and involving a possibility of trouble later on. Steel-shot 
blasting is a very efficient method of cleaning the surface. 

3 -iU K - 

* M- 

FIG. 71. Two-tank system. 

Oil Filters and Settling Tanks. When a turbine is in normal 
operation and has been thoroughly cleaned, the amount of 
impurities that get mixed with the oil is usually small, and 
as far as the oil-circulation system itself is concerned, the only 
precautions as regards filtering may be confined to a good sieve 
in the oil-return tank, a cylindrical strainer on the pump suction 
pipe, or a set of gauze strainers. Ample capacity of the oil tanks 
is always a desirable feature, leading to longer life of the oil 
and also giving the impurities and water a chance to separate out. 


A special design of separating tanks, under the name of the 
" two-tank system," is used in a great many turbine ships 
(see Fig. 71). 

The two tanks (1) shown are not intended to be used concur- 
rently. The oil is allowed to rest in one of them for a certain 
period, while the oil circulation takes place through the other. 
When the oil has " rested " a sufficient length of time to ensure 
complete separation from water and other impurities, the large 
drain cock (2) placed at the lowest part of the tank is opened, 
and the water, dirt, and sludge are drained away until pure oil 
appears. Means should be provided to show clearly the amount 
of water in the oil, and for this purpose a glass-sided box (3) is 
placed at one end of the tank in preference to the ordinary gauge 
glass; a strip of }-in. steel plate should be placed at the ends 
of the box to slide in a groove, the idea being to prevent breakages 
of the glass and, by lifting the steel sheet, to enable one to see 
the amount of water separated out. An air pipe (4), as shown in 
the drawing, should be fitted to the highest point of the tank and 
led to the necessary height. The oil passes a filter (5) on its way 
to the oil pump. 

In cases where a large proportion of water finds its way into the 
oil, a heater might be fitted in the return pipe to raise the tem- 
perature of the oil to about 150F. This will result in immediate 
separation of all water and foreign matter as soon as the oil enters 
the suction tank, the oil rising quickly to the top, and the sepa- 
rated matter remaining at the bottom. To facilitate separation, 
the return pipe should go almost to the bottom of the tank and 
deliver the oil in a downward direction. The tanks should be 
tilted; the suction of the pump should be placed as high as possi- 
ble, the opening of the pipe to be directed upward if possible. 
The net storage capacity of the tank is, of course, the capacity 
above this level. On leaving the tank the oil is sucked through 
a filter consisting of three or four separate layers of gauze of, say, 
24 mesh to the inch, the uppermost layer consisting of two 
sheets of gauze with a sheet of cheesecloth between them. The 
bottom of the filter forms a convenient receptacle for any dirt 
that may have been carried as far as this point, the dirt dropping 
downward from the filter gauze. 

In small and medium-size turbine plants ashore, where, as a 
rule, each turbine has its own separate oiling system, the two- 


tank system has only rarely been employed. The oil circulates 
continuously and gets little rest when the turbine is in operation. 
In such plants it is good practice to remove daily from 3 to 6 gal. 
of oil from each turbine unit, treating this oil in a steam-heated 
separating tank and filter. The purified oil should be returned 
to the circulation system at the same time that a corresponding 
quantity is drawn off for treatment. In this way the vitality of 
the oil can be maintained at a high standard. If the oil-tank 
capacity is small, it is particularly desirable to follow this 

In large turbine power stations consisting of several units it is 
often desirable to have a separate plant for supplying the oil to 
the various turbines and for cooling and purifying the return oil. 
There are several designs of such plants, but common to them all 
is the feature that a portion of the oil is by-passed through a filter, 
while the main flow of oil is only strained and cooled, not filtered. 
The oil coolers, oil filters, and oil tanks are all made up from 
several identical units, so that the necessary cleaning and inspec- 
tion can be made while the plant is in operation and without 
disturbing the normal operation of the oil plant. 

Centrifugal Purification. The best and most efficient method 
of cleaning turbine oil is, however, by centrifugal purification. 
(For description of the purifier see page 598.) 

The purifier protects steam turbines automatically and con- 
tinuously, removing all water and impurities from the oil in 
circulation; it maintains the lubricating and cooling properties 
of the oil almost indefinitely and ensures maximum efficiency 
of the oil cooler by keeping the tubes free from deposit and 

Keeping the oil dry and free from metallic impurities is 
obviously of the very greatest importance, and particularly so in 
connection with geared turbines, in which the problem of how to 
avoid wear of the gears is still a troublesome one. 

The centrifugal purifier also helps to keep the contents of 
petroleum acids at a low figure, especially if hot water of condensa- 
tion is added to the oil entering the purifier. 

This washing of the oil is also beneficial if, as is sometimes the 
case, sea water enters the oil. 

The purifier is preferably installed so that from 5 to 10 per 
cent of the actual flow of oil through the lubricating system is 


by-passed through the purifier, the impurities being continuously 

Oil Consumption. The make-up for lost oil due to leakage 
and atomization there is very little evaporation amounts to 
from 1 pt. to 4 gal. per week per turbine unit, depending upon 
the size and operating conditions. The average make-up for a 
1,000-kw. turbine is about 1 to 1^ gal. per week. 

Acquired Impurities. During the passage of the oil through 
the entire circulation system it picks up more or less water, air, 
iron oxides, and other impurities and, when passing through the 
main bearings, gets intimately churned together with these 
foreign matters ; the result is that, owing to the high temperature 
and the great surface speed of the revolving shaft, the oil gradu- 
ally breaks down. 

When ordinary oils, not specially manufactured for turbine 
use, are employed they may not have a life of more than a few 
months, whereas high-quality turbine oil may last under normal 
conditions 10,000 working hours or more and 3,000 working 
hours under very unfavorable conditions; also, the margin of 
safety will be considerably greater when using the best possible 
oils. Whereas all oils, even the best, are affected in time, unsuit- 
able oils will sooner show the signs of breaking down, which arc 
(1) darkening in color, (2) increased specific gravity, (3) increased 
viscosity, (4) increased acidity, and (5) the throwing down of 
various kinds of deposits. 

The first three effects cannot be said to be detrimental except 
that they are the " signs of warning" that the oil is breaking down. 
As regards the acidity, the acid produced in the oil is the result 
of oxidation and is a petroleum acid which must not be confused 
with sulphuric acid, which is sometimes found in mineral oils 
that have been treated with this acid during their manufacture. 
Petroleum acids do not attack the metals ordinarily used in the 
construction of the circulation system, but they do slowly dis- 
solve zinc or alloys consisting largely of that metal. Increased 
acidity can always be taken as a guide to judge how far the oil 
has suffered, and when it gets in the neighborhood of 0.3 per 
cent in terms of 80s, steps should be taken to prevent this limit's 
being exceeded, by renewing either part of the oil or all of it, as 
the circumstances may seem to justify. When the acidity of an 
oil is below 0.03 per cent it is generally considered by chemists 


to be "free from acid"; good turbine oils often contain less than 
0.008 per cent of acidity when new. 


Deposits may form even where the best oils are in use, although 
always in very much smaller quantities than where unsuitable 
oils are employed. Naturally, it is the constant aim and endeavor 
of the oil manufacturer to produce oils that possess as great a 
resistance as possible against the oxidizing and emulsifying effect 
of the impurities, etc. 

The principal causes of deposit, apart from the quality of the 
oil, are 

1. Water. 

2. Solid impurities. 

3. Air. 

4. Electric action. 

5. Adding new oil. 

1. Water. Water has an emulsifying effect on the oil, par- 
ticularly if it contains impurities, whether in solution or in suspen- 
sion. Where considerable quantities of water leak into the 
system, and emulsification takes place, the oil becomes yellow or 
brownish yellow in color; and if a sample is taken out and heated, 
it will separate into clean oil at the top, more or less milky water 
at the bottom, and a spongy sludge separating the oil and water. 
If the oil and the water are removed, the spongy emulsion, which 
varies in color from gray to brown, will be found to contain 
from 15 to 35 per cent of oil and to consist of numerous exceed- 
ingly thin films of oxidized matter surrounding small drops of 
water; in fact, the sludge when freed from oil consists of about 
99 per cent water by weight and 1 per cent of exceedingly thin 
films. On analysis these films have been found to be composed 
of a chemical combination of petroleum acids, produced by 
decomposition of the oil, and rust (iron oxides), which is found 
throughout the system. 

The nature of the sludge in the oil produced by the water is 
most objectionable, as it tends to clog the oil strainers, /nl inlets 
to the bearings, and oil inlet to the governor. Furthermore, the 
oil pressure may be reduced, owing to the oil pump's not deliver- 
ing the requisite quantity of oil because of the partial choking of 
the pump strainers. The chief source of water's getting into the 


oil is usually the gland packings; water may also leak into the oil 
in the oil cooler or in the bearings (when water cooled). On sea- 
going ships a leakage of cooling water into the oil can be detected 
at once by taste, the cooling water being salt. If one could 
always be sure of the steam passing the glands producing an 
absolute soft water of condensation free from boiler salts, it 
would be an easy matter by analysis to determine whether the 
water leaking into the oil was from the glands or from the cooler 
or partly from one, partly from the other source. But when 
the boilers prime, such analysis becomes almost useless, for 
obvious reasons. Determining the degree of hardness of the 
water drawn away from the oil in the system is quite misleading, 
as the acidity of the water washed out of the oil upsets the 
titration test. Evaporating the water to dryness, ignoring the 
percentage of metallic salts iron, copper, etc. that the water 
has dissolved from the oil pipes, and comparing the grains of 
salts remaining with the results when evaporating a similar 
volume of cooling water are about the only reasonably accurate 
chemical method of forming an idea as to where the leakage 

Mechanically, it is often possible sometimes even quite easy- 
to locate the leakage. 

Where leakage of water into the oil system cannot very well be 
avoided, a " water leg" consisting of at least 4 ft. of vertical 
pipe 2^2 to 4 in. in diameter fitted to the bottom of one of 
the oil tanks may do great service, as it will catch the fine drops 
or particles of water circulating with the oil; and once a particle 
is caught in the leg it cannot again rise and mix with the oil; it 
goes to the bottom of the leg, which should be drained twice 
every 24 hr. Strict instructions should be given that the drain 
cocks in the oil tank or tanks should be opened twice every 
24 hr. and every time the turbine is about to start up after a rest ; 
the drains should be kept open until clean oil appears. 

Turbine oils are affected by water if it contains boiler salts in 
solution, more than by clean water, and certain boiler compounds 
have a strong emulsifying effect, but the greatest effect seems to 
be produced by iron salts in solution. The water cannot help 
dissolving some of the iron during its rapid flow through the oil 
pipes, hence the desirability of using copper oil pipes in preference 
to iron pipes; copper is little attacked by water, and a copper 


solution has only a slight emulsifying effect on the oil as compared 
with the effect of an iron solution. 

2. Solid Impurities. The disintegrating effect on the oil 
caused by finely suspended solid impurities, such as fine rust and 
molder's sand, is very marked. The oil darkens considerably 
in color; the acidity increases rapidly; the oil assumes a " burnt" 
odor; a slimy dark-colored deposit develops and lodges particu- 
larly in the oil cooler. If, furthermore, there is a leakage, how- 
ever slight, of water into the oil system, the oil may get badly 
emulsified, much more than would be the case with water alone, 
as the oil is in a weakened condition due to the oxidizing effect 
of the solid impurities. This will explain why, when a new 
turbine is being started up for the first time, emulsification of the 
oil may occur even if the oil is of good quality. 

Where the inside of the oil tank is painted, emulsification and 
breakdown of the oil usually occur, as there exist hardly any 
paints that are "oilproof" under the exacting conditions pre- 
vailing in turbine practice. The advisability of changing the 
initial charge of oil will, in view of what is said above, now be 
fully understood, the effect being that the entire system gets 
thoroughly cleaned and that the fresh charge of oil will have 
very much better conditions to work under. 

3. Air. The circulating oil always contains more or less air; 
and when the temperature is above normal, say more than 
140F., this air has a tendency to oxidize the oil, a tendency that 
increases rapidly with increasing temperatures. This effect will 
be better realized when one considers that the oil film in the bear- 
ings is very thin and that the air is present in exceedingly fine 
bubbles, which are intimately mixed with the oil. The result is 
that the oil darkens in color, increases in acidity, and in extreme 
cases a black, carbonaceous deposit develops, which is exceed- 
ingly dangerous, as it may choke the oil inlets to the bearings 
and cause sluggish working of the governor gear or may even 
cause it to stick, putting the governor out of action. 

Another effect of air in the oil shows itself only when an abnor- 
mal amount is present; the effect is known as " fuming." Fumes 
issue from the main bearings and oil tank, notwithstanding that 
the bearing temperatures are quite normal; the fumes may be 
drawn into the generator windings and cause disastrous results. 
The cause of the " fumes" is that the fine air bubbles, with which 


the oil is heavily charged, burst in the bearing cavities and in the 
oil tank, producing a very fine spray of oil which oozes out in 
the form of a mist the oil "fumes." The oil will be found creep- 
ing all over the outside of the bearings and turbine bedplate, 
forming a very thin film, and the loss of oil may be quite con- 
siderable several gallons per 24 hr. The remedy is to prevent, 
as far as possible, the oil from getting churned together with the 
air. Perhaps the churning takes place between oil throwers and 
baffle plates inside the bearings, or the oil gets violently disturbed 
in the sight-feed arrangements in the return pipes or where the 
return branch pipes join the main return pipe, etc. If the spray 
is formed inside the bearings, these should be ventilated, a large 
pipe connection being taken from the air space in the bearing 
cavities to the oil-return tank. The fumes will then go through 
these pipes instead of oozing out of the bearing ends; sometimes 
enlarging the oil-return pipes will overcome the trouble. The main 
return-oil tank should always have a vent pipe, at least 1 in. in 
diameter, to prevent accumulation of oil fumes in this tank and 
in the return-oil pipes. Frothing may also occur temporarily 
when a considerable percentage, say 50 per cent, of the oil in 
circulation is renewed at one time. New oil should always be 
added in small quantities at a time. 

4. Electric Action. If in the case of the electric generator 
there is a slight leakage of electric current from the generator 
(direct-current generator), or if the magnetic field is out of balance 
(alternating-current generator) and produces induced currents 
in the turbine shaft, the result is that an electric current passes 
through the shaft down through one of the main bearings, through 
the bedplate, and up through another main bearing back into the 
shaft. The effect on the oil is that it quickly darkens in color, 
increases in acidity, and throws down a deposit which coats all 
parts of the turbine with which it comes in contact, lodging 
particularly in the oil cooler. The deposit is of a fairly hard, 
brittle nature and of dark chocolate color; it is exceedingly diffi- 
cult to remove and therefore very objectionable. The remedy is 
completely to insulate electrically one of the main bearings from 
the turbine bedplate, including the connections between the oil 
pipes and that particular bearing. This insulation will prevent 
the formation of an electrical current, and consequently the 
formation of deposits will cease. 


On rare occasions local galvanic currents may cause corrosion 
of the oil tubes in the oil cooler or of the turbine shaft and bear- 
ings and even in the governor, causing the oil-operated piston to 
stick, or may eat away the sharp edges of the pilot valve. 

5. Adding New Oil. Where practically no water enters the 
circulation system, and where practically no waste or leakage of 
oil occurs, so that the amount of new oil added to the system per 
week is only very small, the oil in time becomes very dark in 
color, and the acidity increases considerably. In such cases it 
has been found that when new oil is added a dark deposit is 
thrown down throughout the system, owing to the action of the 
old oil on the new, and this is particularly the case with heavy- 
viscosity oils rather than with light oils. 

Speaking generally, deposits are always inclined to accumulate 
in the most dangerous places, such as the oil pipes leading from 
the main oil pipe into the main bearings. A partial choking of 
the oil inlet would reduce the oil feed ; the bearing would heat up 
quickly; and if not observed in time the bearing surfaces would 
with all certainty be destroyed, which might have very serious 
consequences, owing to the high speed at which all turbines 
operate and particularly so on account of the time that it takes 
half an hour or more for the turbine to come to rest from full 
speed. If deposits get into the oil pipe feeding the governor gear, 
the governor may fail to act, and consequently the turbine would 
either gradually slow down or increase in speed much above the 
normal speed. The parts inside the governor gear in contact 
with the oil are very sensitive with small clearances and the 
oil must be absolutely clean and good in order to make the 
parts work smoothly. 

Example 1 : 1,000-fcw. Turbine, 3,000 r.p.m. 

Temperature of oil leaving bearings ................. 120F. 

Temperature of oil leaving cooler ............ ...... 110F. 

Quantity of oil in circulation ....................... 60 gal. 

The oil cooler contained 100 copper pipes 21 mm. in 
and 1 m. long, the oil being sucked from the cooler, with the result 
that a slight amount of water was always leaking into the oil in 
the cooler owing to the thin copper tubes 1 not keeping quite 
watertight in the end plates, The cooling water was taken from 


a brook; it was practically soft water, and for several years no 
trouble had been experienced. The oil in use was similar to 
circulation oil 1 (page 408) and gave complete satisfaction. Sud- 
denly trouble began. An emulsified sludge was formed through- 
out the oil system, and the bearing temperatures increased. It 
was found necessary to change the oil every 3 to 4 weeks, whereas 
previously the oil (without any daily treatment) was renewed 
only every 6 months. 

A thorough examination revealed the fact that the oil cooler 
was leaking and furthermore that the water supply had been 
changed. The water instead of being taken from the brook was 
taken from the coal-washing plant, after indifferent filtration; it 
contained coal dust and was exceptionally hard. An emulsifi- 
cation test with fresh oil, using this water, showed unsatisfactory 
separation and explained the cause of the trouble. 

As a result of the higher bearing temperatures which had 
prevailed for several months a large amount of sludge had settled 
in the oil cooler and gradually baked into a fairly hard deposit 
which almost choked the cooler. 

Example 2: 3,000-fcw. Turbogenerator. 

Oil temperature of bearings 120 to 130F. 

Quantity of oil in circulation 120 gal. 

Rate of circulation Exceptionally rapid 

A certain amount of sludge was continuously developed in the 
oil system and settled at the bottom of the turbine-bed chamber 
in the form of an oily sludge. Two samples were drawn at an 
interval of 7 weeks and analyzed as shown in table on page 231. 

It will be seen that the color of the oil, which when the oil 
is new is about 35, has darkened considerably, also that the acidity 
of the oil and sludge has increased between the dates of taking 
the two samples. The cause of the deposit is emulsification 
and oxidation of the oil, brought about by the rapid circulation 
(aeration of the oil) and also the very small volume of oil in 
circulation. Only a small amount of water was leaking into 
the system, but owing to the very rapid circulation of the oil the 
water was never given a chance to separate out. 

Example 3. Four large turbines were using an exceptionally 
heavy turbine oil similar to circulation oil 3. The bearing tem- 
peratures were high from 150 to 165F, and the oil coolers 



were constantly filling up with a thick sludge. An investiga- 
tion proved that the boilers were priming. The boiler salts 
carried over with the steam found their way through the turbine 
glands into the bearings, contaminating the oil and causing the 
sludge. Owing to the fact that the oil was far too viscous for 
the conditions, the cumulative effect of the water charged with 
boiler salts was very troublesome. 

A change in grade of oil to a light-viscosity oil similar to 
circulation oil 1 was made, with the result that the bearing 

Oily sludge and its contents of sludge and oil 

Sample 1, 
per cent 

Sample 2, 
per cent 

Oily sludge: 










45 1 



34 2 

34 2 

Volatile matter insoluble in petroleum spirit. . 
Ash (containing oxides of iron, silica, and lead) 
Petroleum acids 

1 389 as 


1 598 as 



SO 3 

SO 3 

Color, Lovibond ^ in. cell 



temperatures were reduced to 120 to 130F. ; and at the same time 
an efficient system of daily treatment of the oil in the turbine 
was instituted. It was then found that very little sludge formed 
in the system and that the little that did form was largely 
removed from the oil by the process of daily treatment. 

Example 4. That an admixture of fixed oil, whether vegetable 
or animal, quickly causes trouble when water is present is obvious 
and usually very soon detected. The following example is of 
interest. in this connection. 

A new grade of turbine oil was tried on board a larg$ turbine 
steamer, the entire system being cleaned out and filled with it. 
On the first trip the new oil became badly emulsified, and the 
chief engineer, complaining bitterly, insisted upon reverting to 
the old. Careful examination proved, however, that there was 


a small percentage of saponifiable matter present in the turbine 
system and in the turbine oil-supply tanks on board the boat; 
and strangely enough the percentage of saponifiable matter, 
although very small, was greater in the oil circulating in the 
turbine than in the oil in the supply tanks. It was evident that 
some compounded marine-engine oil had been "accidentally" 
added to the system, and evidently a slightly greater proportion 
had been added to the turbine system than to the supply tanks. 

In connection with marine steam turbines great care must 
be exercised to prevent contamination with marine-engine oils, 
which are always compounded with vegetable or animal oils. 
This point must be particularly watched in case of large warships, 
where oil is pumped on board through a flexible hose; a separate 
line must be used for turbine oil. 

Example 5. In a large turbine shortly after erection the 
bearing temperatures began to rise, and a tenacious emulsified 
sludge developed throughout the system. It was found that 
the water-softening plant for treating the boiler water had not 
been properly looked after, excess soda getting into the boilers. 
Priming of the boilers carried soda into the turbine, and through 
the glands it finally reached the oiling system. The turbine-bed 
chamber was painted with "oilproof " paint, but the soda very 
soon dissolved or destroyed it, and mixing with the water brought 
about the emulsification. 

Example 6: 3,000-&w. Turbine. An oil similar to circulation 
oil 1 and of good quality was in use. Oil temperatures were 
normal, being approximately 120F. The quantity of oil in 
circulation was 60 gal.; it was drawn through the cooler by the 
oil pump, so that it was always under suction. Very little water 
leaked into the oil system, being approximately 1 pt. per 24 hr. ; 
the oil gave excellent results and was renewed only once a year. 
A thin deposit having the following composition developed in the 
oil cooler: 

Per Cent 

Oil with a trace of moisture 46 . 4 

Volatile matter insoluble in petroleum spirit 48. 9 

Fixed carbon and silica. 0.1 

Iron oxide 2.2 

Copper oxide 2.0 

Balance undetermined 0.4 


A sample of the water leaking into the oil system was analyzed 
and found to be very hard, similar to the cooling water. Obvi- 
ously, the cooling water had constantly leaked into the oil system; 
owirig to the small volume of oil in rapid circulation, considerable 
aeration took place, and the combined effect of the air and water 
produced slowly the deposit that was found in the oil cooler. 

This and Example 1 point to the desirability of always having 
the oil under a pressure in the oil cooler higher than the pressure 
of the cooling water. 

Example 7. A 1,500-kw. turbine had for several years been 
using an oil similar to circulation oil 1 with every satisfaction. 

Quantity of oil in circulation 80 gal. 

Bearing temperatures Quite normal 

Suddenly the bearing temperature rose within one week from 
about 110 to 140F. On examination it was found that a thick 
deposit had developed and nearly choked the oil coolers. The 
deposit on analysis gave the following composition : 

Per Cent 

Oil and water 42. 8 

Volatile matter insoluble in petroleum spirit 17.8 

Fixed carbon and silica 1.6 

Iron oxide 36 . 4 

Balance undetermined, containing copper oxide, etc. ... 1.4 

Analysis of the oil showed that it was in very good condition, 
the percentage of petroleum acids being only 0.05 per cent. It 
was somewhat dark in color and heavier in viscosity than the 
fresh oil but nothing to be alarmed about. 

On the oil pipes' being taken apart it was found that during 
5 years' operation the pipes had rusted on the inside, and a por- 
tion of the rust had been either absorbed by the water circulating 
with the oil or circulated in the form of a fine powder. 

As mentioned elsewhere, finely divided iron and iron salts have 
a powerful effect on turbine oils. This explains the formation of 
the sludge which almost put the oil coolers out of action and 
brought about the high bearing temperatures. 

Example 8: 1,700-fcw. Turbogenerator, 3,000 r.p.m. 

Quantity of oil in circulation 60 gal. 

Temperature of oil leaving bearings. . Approximately 150F. 
Temperature of oil leaving cooler 140F. 



Great trouble was experienced in this turbine with oxidation. 
A black brittle deposit developed throughout the system, set- 
tling particularly in the oil cooler and in the oil inlets to the bear- 
ings, also in the governor gear, preventing the governor from 
functioning properly. 


Specific gravity ) 

Open flash point > Unaltered 

Fire point j 

Saybolt viscosity at 104F., seconds 


Petroleum acids as SO 3 , per cent 

Deposit : 

Volatile matter insoluble in petroleum spirit, per cent 

Ash, chiefly iron oxide, per cent 









The analysis given in the preceeding table compares the unused 
oil and the oil after 4 months' use. 

It is obvious that the temperature of the oil in circulation was 
too high and the amount in circulation too small, with the result 
that the oil was quickly oxidized. 

Example 9. A large turbine suddenly developed high bearing 
temperatures, and an investigation proved that the vertical oil 
cooler had become air locked, the upper part of the oil cooler thus 
being put out of action. The obvious remedy was to fit an air- 
vent pipe, leading the air from the uppermost part of the oil cooler 
up to the main oil-return tank. 

Example 10. A 1,000-kw. steam turbine immediately after 
erection was greatly troubled with oil vapors oozing out of the 
turbine bedplate (used as the oil reservoir), which meant not only 
a large waste of oil but also a considerable danger to the generator. 

An investigation proved that the oil-return pipes from the 
bearings, instead of sloping gradually into the bedplate, were 
vertical; the return oil falling into the reservoir caused the oil 
to splash about and form a great deal of oil spray. The return- 
oil pipes were then altered, and the trouble ceased. 

Example 11. A 1,500-kw, steam turbine was greatly troubled 
with oil vapors which evidently emanated from the main bearings, 



and the presence of oil in the generator was clearly visible. 
Everything possible had been tried to stop the vapors emanating 
from the bearings, when on an investigation by an oil expert it 
was found that the return-oil tank had no vent pipe and that the 
fine oil spray developed in the bearings could not pass back into 
the oil tank but simply filled up the oil-return pipes and then had 
to find its way out through the bearing ends. The obvious 
remedy was applied, and the trouble thus overcome. 

Example 12. A 1,500-kw. Howden turbine, 3,000 r.p.m., was 
using an oil similar to circulation oil 2 and of good quality. 
Difficulties were experienced with the oil's " creeping" along the 
turbine shaft and getting into the generator. 

An investigation proved that the oil was unnecessarily viscous 
for the conditions, and an oil similar to circulation oil 1 was 
installed to see whether the change would make any difference. 
Curiously enough, the creeping of the oil entirely disappeared 
without the engineer's being able to offer any definite explanation 
as to the reason why it ceased. At the same time, a remarkable 
difference in the bearing temperatures took place, as shown in the 
following table: 





Bearing 1 



Bearing 2 



Bearing 3. . . .... 



Bearing 4 



Bearing 5 



Bearing 6 



Temperature of inlet oil 



Temperature of outlet oil 



Temperature of inlet cooling water 
Temperature of outlet cooling water. 



Temperature of engine room 



The foregoing figures show clearly the lower bearing tempera- 
tures obtained by using the low-viscosity oil, notwithstanding 
that the supply of town water through the oil cooler was greatly 
decreased when the new oil had been installed; as town water had 


to be paid for, the change in oil brought about a quite considerable 
saving in the water bill. 

NOTE. Where oxidation takes place owing to oil temperatures 1 being 
too high, a change to lighter viscosity oil has often reduced temperatures 
and stopped the oxidation. 

Example 13: 350-fcw. Mixed-pressure Turbine, 3,000 r.p.m. 

Temperature of oil leaving bearings. . 120F. 
Temperature of oil leaving cooler .... 105F. 

Quantity of oil in circulation 60 gal. 

Oil consumption 1 gal. per week 

added to the system 

The turbine was in operation day and night continuously until 
it had to be stopped owing to the breaking down of the armature 
of the generator. Before the turbine was stopped the oil tem- 
peratures had for several weeks been gradually creeping up, for 
some unknown reason. When the turbine was opened up for 
inspection the flexible coupling between the turbine and the gen- 
erator was discovered to be absolutely solid with a black brittle 
carbonaceous deposit, which was also found throughout the 
entire oil system. 

Strangely enough, there was no perceptible wear of any of the 
bearings; the surfaces of the brasses were black and dull, covered 
with a very slight deposit. The oil from the turbine had a 
charred odor and a dark-brown bloom, whereas the bloom of the 
fresh oil was green. It was apparent that a radical change had 
taken place in the oil. The deposit consisted of 

Per Cent 
Oil and volatile matter insoluble in petroleum spirit, with 

a slight percentage of water 77 . 4 

Fixed carbon 2.4 

Iron oxide 10.5 

Copper oxide 8.4 

Undetermined, containing carbonate of magnesium, 

traces of lead, etc 1.3 

The total amount of the deposit was about 25 lb., and a large 
portion of this had undoubtedly been in constant circulation with 
the oil in the form of very fine powder which settled when the 
turbine was stopped. Owing to a fault in the rotor and armature, 
stray currents had passed down through the bearings, oil pipes, 


oil cooler, etc., and had caused the oil to break down, developing 
the deposit. 

The remedy, apart from putting the rotor in order, was to 
insulate the end bearing of the turbine entirely from the bedplate. 
This practice is now followed by a good many turbine builders. 

Example 14. A 1,000-kw. exhaust steam turbogenerator had 
an electric breakdown similar to that of the turbine mentioned in 
Example 13. The oil used underwent a remarkable change in 
the course of one week, becoming changed in color from 35 to 180 
and the acidity increasing from 0.002 to 0.298 per cent; simultane- 
ously, the viscosity increased about 20 per cent. A brownish 
brittle deposit with a lustrous fracture developed throughout this 
system and had the following composition: 

Per Cent 

Water 24.2 

Oil 17.2 

Volatile matter insoluble in petroleum spirit. . . 52.0 

Ash, chiefly iron oxides 4.3 

Petroleum acids 2 . 3 as SO 3 


These turbines are found chiefly in the United States, only a 
few having been installed in England. They operate electric 
generators and are made in sizes from 500 to 20,000 kw., the 
corresponding speeds ranging from 800 down to 720 r.p.m. The 
revolving parts are supported by a combined step-and-guide 
bearing and by upper and middle guide bearings. The middle 
bearing may be left out with smaller machines where the turbine 
shaft is in one piece. 

The step bearing is shown in Fig. 72 and consists of two cast- 
iron blocks, one carried by the end of the shaft, and the other 
held firmly in a horizontal position and so arranged that it can 
be adjusted up and down by a powerful screw. The lower block 
is recessed to about half its diameter, and into this recess oil is 
forced with sufficient pressure to balance the weight of the whole 
revolving element; there are, of course, no oil grooves. The 
amount of oil required is small from 1^ gal. per min. for a 
500-kw. machine to about 6 gal. per min. for an 8,000-kw. 
machine. The oil, after passing between the blocks of the step 
bearing, wells upward, lubricates a guide bearing supported by 
the same casting, and leaves through oil drain (1), 



A carbon packing, prevented from rotation and consisting of 
two sections of rings, each section comprising two rings made up 
from three segments, is fitted above the oil thrower (2), and, in 
order that no oil or air shall enter the turbine chamber above the 
packing, a low steam pressure is maintained between the two 
sections of the packing, just sufficient so that vapor is visible 

FIG. 72. Curtiss step bearing. 

at the outlet of the drain pipe (3). If the flow of oil into the bear- 
ing is too great, the oil overflows into drain (3), mixing with the 
steam; the mixture should be drained into a separate tank with 
baffle plates, in which the water is held back; the recovered oil 
may be allowed to enter the main oil system when entirely freed 
from water. 

The oil pressure required for the step bearing is slightly higher 
than the bearing pressure, ranging from 300 to 800 Ib. per square 
inch, thus producing perfect oil-film lubrication. To start lubrica- 
tion a pressure 25 per cent greater than the normal running pres- 




sure is needed. The film thickness depends upon the flow of oil, 
ranging usually from 0.003 to 0.006 in. 

In some designs a powerful brake bearing is provided which can 
be operated from the outside and can be used to take the whole 
weight of the revolving part in case the step-bearing support 
should fail. In ordinary operation the shoes of this brake will be 
set about 0.01 in. below the brake ring. It is thus in a position 
to receive the revolving part in case 
the step-bearing support should fail. 
Another and more important feature 
of this brake is to stop the machine 
when it is desired to do so. A 5,000- 
kw. machine will run for 4 or 5 hr. after 
the steam has been shut off, unless a 
brake is applied. 

In some cases the step bearings have 
been operated with water instead of oil, 
in which case no packing is necessary, 
the water being allowed to pass up into 
the turbine. The trouble with water 
is that it causes rusting of parts. 
When accidentally the step-bearing oil 
pressure has dropped below the pressure 
required, the bearing surface immedi- 
ately cuts; but the metal is removed 
very slowly, and lubrication is easily 
reestablished when the pressure oil flow 

restored. Precautions are, however, 

FIG. 73. Oil pressure baffler. 

is restored. Precautions are, nowever, taken in the shape of 
accumulators and other auxiliaries necessary for the maintenance 
of a flow of pressure oil to the step bearing. 

The guide bearings are babbitt-lined sleeves, with a clearance 
of 0.0005 in. per inch shaft diameter for the lower and twice this 
clearance in the upper and middle guide bearing. They have 
suitable oil grooves to ensure good oil distribution; the oil is fed 
at the rate of 0.5 to 1.5 gal. per min. per bearing according to size 
and is distributed by gravity from an elevated oil tank or from 
branch pipes from the main pressure system. 

In the latter case, bafflers, as illustrated in Fig. 73, are fitted 
to reduce the oil pressure. The oil is forced to pass through the 
narrow spiral passage formed by the thread, and the longer the 



passage the more is the pressure reduced. These bafflers are 
also placed in the delivery line to the step bearing to reduce the 

FIG. 74. 

FIG. 75. 
FIGS. 74-75. Preventing oil spray from guide bearings. 

pressure and also to reduce the intensity of the pulsations caused 
by the reciprocating main oil pump. When the oil leaves the 


guide bearings it is thrown off into oil troughs, large drain pipes 
guiding it back into the oil reservoir. 

A carbon packing is also fitted between the middle guide bear- 
ing and the turbine; excessive steam admitted to this packing will 
mix with the return oil from the middle guide bearing and should 
be avoided. 

The guide bearings may cause trouble by oil throwing, caused 
by leaky joints (which is easily remedied) or by oil spray sucked 
out from the bearings by the draft created by the rotating 
parts. Deflectors, as shown on page 168, may also be adapted for 
vertical turbines, but the disease is a troublesome one to cure. 
Possibly the oil supply is too great, particularly when the oil is 
introduced under great pressure, or the oil troughs may have 
become obstructed by dirt, which may account for the oil's getting 
into the generator; they, as well as the oil-return pipes, should 
therefore be kept clean. If oil leaks through a porous casting, a 
mixture of litharge and glycerin applied to the points of leakage 
is said to be a remedy. 

To stop the fine oil spray from being carried out from the bear- 
ings, it is necessary to equalize the air pressure outside and within, 
as shown in Figs. 74 and 75 for an upper and middle guide bearing, 
respectively. The pipes (1) are pressure-equalizing pipes, taken 
outside to a point where there is no suction; and, in addition, felt 
rings are fitted, as shown, and prove very effective as long as 
they are not worn too much. The arrangements in Figs. 74 
and 75 were designed by E. D. Dickinson of the General Electric 
Company. The casings are made of sheet iron with riveted 
joints; the felts are fastened by means of metal rings. Where 
bolts are used they should be locked, so that the nuts cannot 
come undone through vibration. 

Oil Distribution. The oil is distributed under pressure to the 
step bearings and guide bearings as described, but the latter are 
sometimes fed from an elevated tank, with an overflow pipe back 
to the oil reservoir. On its way to the bearings the oil pressure is 
reduced by one or more bafflers, so that each bearing gets the right 
amount of oil. The oil is returned by gravity from the bearings 
and passes a filtration and cooling system, in which it is freed from 
water, dirt, and other impurities, before it is circulated afresh. 

The oil reservoirs should preferably be in duplicate and 
operated alternate days. 


When water is used for the step bearing, the oil pumps have 
to supply oil only for the upper and middle bearings, usually by 
way of an elevated tank. The amount of oil going to each 
bearing is regulated by small control valves and sight feeds in 
each line. 

In an installation of one or two units of the same capacity, 
two high-pressure steam-driven pumps supply oil to the step bear- 
ings, and two low-pressure pumps supply the upper and middle 
guide bearings and an accumulator gear, for equalizing variations 
in pressure caused by fluctuations in the speed of the pumps. 
These accumulators in case of failure of pump will keep the 
turbines running for some time and automatically cause reserve 
oil pumps to come into action. The accumulators may be on 
the principle of a heavy weight which is raised or lowered accord- 
ing to the amount of oil "stored" in the accumulator. Air 
chambers have also been used as pressure accumulators and must 
be absolutely airtight. In installations of three units of the 
same capacity three high- and three low-pressure pumps are 
fitted, two sets being sufficient to supply all units. 

In a plant comprising two or more units the starting or stopping 
of a unit means that the amount of oil required is altered; the 
alteration in oil supply is automatically brought about by influ- 
encing the speed of the oil pumps. The latter should run at no 
greater speed than that required to give the necessary oil supply 
plus a margin. If the speed is greater, power is wasted, and an 
excessive oil supply may cause various kinds of trouble, such as 
oil throwing, oil overflow into packing drain pipe, excessive churn- 
ing of the oil in the pumps (causing emulsification when water is 
present), etc. 

The number of gallons of oil in circulation is about 10 per cent 
of the rated kilowatt capacity for turbines of 4,000 kw. or over; 
20 per cent for turbines between 2,000 and 4,000 kw.; and a 
still higher percentage for smaller turbines, being 200 gal. for a 
500-kw. unit. 

Oil. The step-bearing lubrication is not dependent upon the 
viscosity of the oil; the shaft floats on the oil film, whether the 
oil is thick or thin, simply because the oil is introduced at a 
sufficiently high pressure. Some "body" is, however, required 
for lubricating the guide bearings, particularly when there is a 
tendency to vibration. Very low-viscosity oils were at one time 


used for Curtiss turbines, but any leakage is accentuated by 
their use, and more oil spray may be formed in the bearings. The 
oil must, of course, be a circulation oil in order to separate well 
from water and withstand oxidation. Unless the conditions 
specially call for a more viscous oil, circulation oil 1 should be 
recommended in all cases. 


For satisfactory lubrication of steam turbines only three oils 
are required, having approximately the following specifications: 


Specific gravity . 870 

Flash point open 395F. 

Viscosity No. 4 (see page 57) 

t.e., viscosity in centipoises at 50C. : 13 
Setting point 20 to 25F. 


Specific gravity . 875 

Flash point open 410F. 

Viscosity No. 7 (see page 57) 

i.e., viscosity in centipoises at 50C.:26 
Setting point 35 to 40F. 


Specific gravity . 880 

Flash point open 425F. 

Viscosity No. 8 or 9 (see page 57) 

i.e., viscosity in centipoises at 50C.:38 or 56 
Setting point 35 to 40F. 

* All circulation oils must separate rapidly from water, and only a trace of sludge must 
be produced in the emulsification test. 

For Steam Turbines 

Land Turbines. Circulation oil 1 is suitable for the great majority of 
land turbines, including the vertical type of Curtiss turbines. 

During the last ten years or so, turbine builders have gradually realized 
the importance of using a light-viscosity oil for high-speed turbines and have 
designed the lubricating system in such a manner that the pump pressure 
required to operate the governor gear can be obtained, notwithstanding the 
use of a low-viscosity oil. 

The advantages of such an oil as compared with a heavy-viscosity oil are: 

Lower fractional losses (i.e., low bearing temperature). 


Rapid removal of heat from the turbine bearings. 

Rapid cooling of the oil in the coolers. 

Quick separation from water, dirt, and other impurities. 

Longer life of the oil. 

Greater freedom from trouble. 

When circulation oil 1 is not viscous enough to give the pump pressure 
required for the governor gear, circulation oil 2 or even 3 (in very special 
cases) must be used. 

Marine Turbines. Marine turbines operate at lower speeds than land 
turbines and with higher bearing pressures. A heavier viscosity oil is 
therefore required, and circulation oil 2 will generally be found to be the 
correct grade. 

Geared Turbine* (Land and Marine). The lubricating system for the gears 
should preferably be separate and distinct from the lubricating system serv- 
ing the turbine bearings, as the conditions of service are entirely different, 
and frequently circulation oil 3 will be found best for the turbine gears. 
Only in rare cases will this oil be the most suitable one for the turbine bear- 
ings, as its use frequently will mean: 

High frictional losses (i.e., high bearing temperatures). 

Slow separation from water, dirt, and other impurities. 

Rapid oxidation of the oil and the development of objectionable deposits 
in the circulation system. 

For turbine bearings in geared turbines, when the lubricating system is 
separate from that serving the gears, a lighter viscosity oil either circula- 
tion oil 1 or circulation oil 2 should preferably be used. 



The parts to lubricate are the main bearings, the crankpin, 
the crosshead and guide, the eccentric straps and sheaves, the 
valve motion, and the governor. 

Main Bearings. In small engines these bearings are siphon 
oiled; in larger engines they are ring oiled or are oiled from a 
circulation-oiling system. 

Crankpin. The crankpin in most engines is oiled by the banjo 
system, the oil being delivered into the banjo either by a sight- 
feed drop oiler or by a pipe from the circulation-oiling system. 

FIG. 76A FIG. 76C FIG. 76B 

FIG. 76. Crosshead and guide lubrication. 

Crosshead and Guide. Lubrication of the crosshead and guide 
may be accomplished as shown in Fig. 76A. Oiler (1) lubricates 
the top slipper; oiler (2) supplies the crosshead, and the oil leav- 
ing these points finally reaches the bottom guide, being retained 
by the splash guards (3). The crosshead pin may also He lubri- 
cated through holes drilled as shown in Fig. 765; the oiler (4) 
feeds oil to the wiper (5) which is fixed on the crosshead and 
delivers the oil to the crosshead pin. 



The lower crosshead slipper is preferably fitted with a comb, 
shown in detail in Fig. 76C, which touches the guide with 
a slight pressure and assists in spreading the oil all over the 
guide; this arrangement is also used to advantage on vertical 

Figure 76 shows a bored guide, which is now commonly used for 
stationary steam engines and which gives greater satisfaction 
than flat guides give. Great accuracy is more easily obtained 
when the surfaces can be bored and turned than when they have 
to be planed. 

In the best constructions the lower guide is drilled so that oil 
from the end wells continuously flows along the horizontal pas- 
sages and up through these 
holes, being distributed by 
means of short transversal oil 
grooves. In the absence of 
this arrangement Fig. 77 shows 
suitable grooving of the bottom 
FIG. 77. Oil grooving of bottom guide guide shoe and chamfered edges 

at either end instead of combs. 

There is a growing tendency to construct stationary steam 
engines, whether vertical or horizontal, with a gravity circulation- 
oiling system, consisting of a pump, top and bottom tank, dis- 
tributing pipes, return pipes, and a strainer or filter in the circuit, 
e.g., the filter (Fig. 222, page 594). This system entails many 
advantages over the ordinary method of distribution, such as 
greater certainty of the oil's reaching every part and greater ease 
in controlling the oil supply; greater margin. of safety in opera- 
tion; lower friction brought about by an abundant supply of 
lower viscosity oil; and an appreciable reduction in oil consump- 
tion, when care is taken to avoid leakage throughout the system. 
The oil wastage in gallons per month will with a good system 
range between 1 and 2 per cent of the engine horsepower. 

The governor, valve motion, etc., are usually hand oiled, but in 
large engines small sight-feed drop oilers are employed for the 
most important parts. 

Bearing Oils. The bearing oils used for external lubrication of 
stationary steam engines are usually straight mineral oils, as they 
come in contact with more or less water of condensation from the 
glands, which would emulsify compounded oils. 



For the crankpins and main bearings, when they do not form 
part of a circulation system, and when they are heavily loaded or 
in bad mechanical condition, compounded engine oils, such as 
one of the marine-engine oils (page 267), are sometimes required 
to keep them "cool." In extreme cases, castor oil has been used 
with great success, but its use should be discouraged on account 
of its tendency to gum. Castor oil is often resorted to in case of 
trouble and is allowed to remain in use instead of correcting the 
mechanical defect and introducing a proper grade of engine oil, 
which will on an average reduce the frictional temperature 50 per 
cent, as shown in the following example, which is typical. 

On the main bearings of a steam engine where castor oil was 
fed through an oil-circulation system, the rise in temperature of 
the bearings above room was 17F. By gradually introducing 
an oil like marine-engine oil 1 (called "X") the frictional tem- 
perature was reduced as follows: 

Grade of oil 

Temperature, degrees Fahrenheit 




Pure castor oil. . . 






90% castor + 10% X 

80 % castor + 20 % X 

60% castor + 40% X 

Pure X oil 

This shows a decrease in the frictional temperature of 53 per 
cent. In changing over from castor oil or any other vegetable 
or animal oil to an oil largely mineral in character, it is necessary 
to exercise great care and make the change gradually, as the 
deposits that have accumulated from such oils are loosened and, 
if loosened too quickly, cause trouble. The deposits when 
loosened gradually are caught in the strainers of the oil pump and 
should be removed as they appear. 

It is not unusual to find steam-cylinder oil in use on guides or 
mixed with the engine oil. This is bad practice, as the great 
viscosity of the cylinder oil causes great friction and high tem- 
peratures; it would be better to introduce a marine-engine oil 
on guides inclined to be troublesome, assuming that they can- 
not be made to run cool on the ordinary engine oil. 



On very large, long-stroke engines with open guides and tail- 
rod supports, the engine oil may be so wasteful in splashing away 
that the use of cylinder oil may be justified. The difficulty with 
splashing from crankpins in long-stroke engines and providing 
proper splash guards has in some cases prompted the use of crank- 
pin grease, usually a white grease, in place of oil. 

In large crankpin bearings or main bearings on slow-speed 
engines, whether grease or oil is employed, oil grooves are some- 

FIG. 78. Oil grooving a main bearing. 

times an advantage when the engine always runs in one direction. 
Fig. 78 shows the proper way of making the oil grooves in the four 
parts of a main bearing; the straight oil grooves and chamfered 
edges collect and feed the oil along their entire length. The 
bearing pressure is constantly squeezing the oil from the center 
toward the edges of the brasses, but the curved grooves help to 
conduct it back toward the center of the bearing. 

When grooving bearings, an important rule is to groove only 
one, not both, of the surfaces, and the grooving should preferably 



be done in the female, or enveloping, surface; for example, the 
bearing surfaces of the connecting-rod brasses are grooved, not 
the crankpin itself (see Fig. 20, page 121). 

As exceptions to this rule note the grooves in Fig. 77 and the 
distributing oil grooves in long spindle bearings for machine 
tools (Fig. 114, page 304). 

For Stationary Open-type Steam Engines 


at 50C. 

System, horsepower 

Circulation oiling 

Drop feed 

Bearing oil 2*. 


18, 20 


Below 250 
250 to 400 
Above 400 
For special cases 

To be used only wh 
subjected to abnor 
are in bad conditio 

Below 100 
100 to 250 
250 to 500 
Above 500 

ere bearings are 
mal pressures or 
n mechanically 

Bearing oil 3 

Bearing oil 4 

Bearing oil 5, 6 

Marine-engine oil 1 and 
marine-engine oil 2. . 

c For bearing oils, see p. 135. 



The vertical, high-speed, enclosed type of steam engine has 
been much developed in England, the engines ranging in size 
from 10 to 2,500 hp., with corresponding speeds of 800 down to 
250 r.p.m. 

The horizontal, high-speed, enclosed-type steam engine has 
come into favor in America for small powers. Both the vertical 
and the horizontal types may be lubricated by the force-feed 
circulation system or the splash-oiling system. 

Force -feed Circulation. Figure 79 illustrates a typical force- 
feed circulation system. The oil pump (1) sucks the oil from 
the oil reservoir and delivers it at 5 to 15 Ib. pressure per square 
inch through pipes (2) into the main bearings. The crankshaft 
is hollow, and the oil is forced from the main bearings into the 
shaft and through oil passages into the eccentric sheaves and 
crankpins, whence it reaches the crosshead bearings through 
passages in the connecting rods or tubes attached thereto. The 
oil leaving the crossheads splashes on the crosshead guides and 
drops back into the crank chamber. It then flows to the oil 
reservoir and reenters the oil pump through a strainer, thus 
completing the circuit. In large engines the guides are fed with 
a direct supply of oil from the main distributing pipe. 

An adjustable oil-relief valve (not shown) is fitted and allows 
a portion of the oil to overflow back into the oil reservoir. In 
this way the oil pressure may be adjusted within certain limits. 
The oil pump should be of ample capacity so that the pump 
pressure, by means of the adjustable relief valve, can be kept at 
any desired point. Too small an oil pump or slack bearings 
decrease the oil pressure or make it necessary to use exceedingly 
viscous oils, which result in unnecessarily high friction losses. 

The oil pump should be placed with its suction strainer elevated 
to leave room below for water to accumulate. Otherwise, water 




is drawn with the oil into the pump and forced through the bear- 
ings, tending to emulsify the oil. Water gets into the crank 
chamber, owing to the presence of ill-fitting glands or "scored" 
rods. Where the rods enter the crank-chamber top, scrapers 

FIG. 79. Force-feed circulation. 

are preferable to glands with soft packing. The oil which is 
carried up from the crank chamber and scraped off, together with 
the water, should be drained to an oil separator outside the 
crank chamber or treated in a steam-heated settling tank to 
recover as much oil as possible. 



Metallic packings are preferable to soft packing in these 
engines, as there is less danger of scoring the rods than with 
soft packing, which is easily screwed up too tight. Once a rod 
is scored, it is impossible to prevent water from traveling through 
the "ridges" down into the crank chamber. 

Slightly superheated steam is an advantage, as less condensa- 
tion occurs in the cylinders; therefore less water finds its way 
through the glands. 

FIG. 80A. FIG. 80. 

Oil and water separator. 

The crank chamber should be systematically drained at suit- 
able intervals. 

A drain cock of preferably not less than 1^-in. bore should be 
fitted at the lowest point in the crank chamber; and if the water 
can be drained off while the engine is running, this should be 
done at frequent intervals. Where the draining cannot be accom- 
plished while the engine is running, it should be done before start- 
ing up, every time that the engine has had a rest. 

When the engine is supplied with wet steam, it is difficult to 
prevent an excessive amount of water from getting into the crank 
chamber, unless the rods and oil scrapers are in perfect condition. 



When this is not the case the piston and valve rods are constantly 
splashed with oil which is carried up through the scrapers. 
Accordingly, a large amount of a mixture of oil and water is 
constantly scraped off. 

Some engines have holes in the crank-chamber top which allow 
the water and oil to drain straight into the crank chamber; obvi- 
ously this is bad practice. Other engines have an automatic 




FIG. 81. Water-drainage tank. 

separator, as shown, mounted on the engine, in Fig. 80A and in 
detail in Fig. 805. The drain pipe (1) from the crank-chamber 
top enters the separator at the side; the oil rises to the surface 
and overflows through the adjustable pipe (2) back into the 
crank chamber; the water flows below a baffle and leaves the 
separator through the drain pipe (3). 

As to the water which drains into the crank chamber, Fig. 81 
shows a useful arrangement. From the lowest point in the crank 
chamber, whether this be at the end or in the middle, a pipe (1) 
is connected to a tank (2) which acts in very much the same way 
as does the "water leg" for turbines. Once water gets into the 



tank, it cannot reenter the crank chamber. Accumulation of 
water should be drained out periodically. 

Every plant should have arrangements for treating daily a 
portion of the oil in circulation, to free it from water, sludge, and 
impurities and so maintain its vitality. This system of daily 
treatment is mentioned (pages 120 and 223) under " Steam 

In large engines it may be necessary to cool the oil to keep its 
temperature below 140F.; a cooling coil made of seamless tube 

1 Scraper Gland 

2 Partition 

3 Piston Rod 

FIG. 82. Piston-rod scraper gland. 

immersed in the oil reservoir is usually all that is required. 
Practically all that is said regarding oil in connection with steam 
turbines applies also to forced-lubrication steam engines and will 
therefore not be repeated here. 

The oil-pressure gauge should be watched regularly. If the 
oil pressure gradually declines, the cause may be bearings requir- 
ing adjustment, emulsification of the oil, or choking of the filter. 

Some engines have double strainers so arranged that either 
can be removed for cleaning without disturbing the action of the 
oil pump. It is of great importance that the strainers be kept 
clean and free from sludge or dirt. 

In horizontal engines it is difficult to prevent oil from splashing 
on to the piston rod and getting into the piston-rod packing; 
with saturated steam this is not a serious matter, but with super- 
heated steam the oil carbonizes in the packing. Figure 82 shows 
a special scraper gland fitted round the piston* rod and fixed in a 


partition. This arrangement is used in some large uniflow 
engines and has proved very effective. 

Mutton cloths should be used for wiping the crank chamber 
when cleaning not cotton waste, which often will cause trouble, 
as the fluffy fibers stick to the surfaces, are afterward carried 
with the oil to the pump, and may choke the strainers. 

The advantages of forced lubrication over the ordinary meth- 
ods are many. The lubrication is entirely self-contained. The 
engines, with correct bearing adjustment and oil pressure, operate 
noiselessly and will run for years, practically without wear, owing 
to the perfect film formation. As the engines are double acting, 
the relaxation of pressure on the upstroke of the pistons gives 
the oil a chance to force itself thoroughly in between the rubbing 
surfaces, forming an excellent cushion for the next stroke. In 
fact, engines may be run with the bearings rather slack and yet 
without noise; there is not sufficient time during a single stroke 
to squeeze the oil film out, particularly if the oil has a high visco- 
sity. If an engine uses an oil too low in viscosity, it is inclined to 
run noisily, and the oil in circulation becomes very warm; the 
introduction of the correct-viscosity oil will reduce the tempera- 
ture and give a sweeter running engine. 

It is in the author's opinion good practice to run with rather 
small bearing clearances and low- viscosity oils; such oils give 
less friction, lower temperatures, separate more easily from water, 
etc., and last longer than viscous oils. 

Forced-feed circulation, when properly arranged, is a very eco- 
nomical oiling system; the consumption of crank-chamber oil 
ranges from 0.05 to 3.0 g. per brake horsepower hour, the normal 
average being 1.0 g. per brake horsepower hour. The consump- 
tion is highest for smaller engines and when a great deal of water 
gets into the oil. 

Grades of Oil. The same oils as are used for steam turbines 
should also be used for forced-lubrication steam engines, and for 
normal conditions they may be recommended as follows: 

For Forced-lubrication Steam Engines 
Circulation Oil 1.* For engines below 150 hp. 
Circulation Oil 2. For engines from 150 to 400 hp. 
Circulation Oil 3. For engines above 400 hp. 

* For circulation oils, see p. 243. 


NOTE. Certain makes of engines operate with unusually large bearing 
clearances; others have unusually stout connections between the cylinders 
and the crank chamber, so that a large amount of heat is carried down 
from the cylinders into the crank chamber. In either case, circulation oil 2 
must be used for engines below 250 hp., and circulation oil 3 for engines 
above 250 hp. 

Splash Oiling. On account of its simplicity and low cost this 
system is used to some extent on small horizontal engines of 
American make, but it is used chiefly for vertical single-acting 
engines, like the Westinghouse engine (United States) and the 
Willans central-valve engine (England), the former being made in 
all sizes up to 200 hp., the latter in sizes up to 1,500 hp. Splash 
oiling is rarely used for vertical double-acting engines. 

The crank chamber is filled with water and oil to a level about 
% in. below the underside of the crankshaft. The cranks dip 
into the bath and splash the oil to the crankshaft bearings, crank- 
pins, eccentrics, and pistons. When the engine is to be started 
with a new "bath" after the chamber is thoroughly cleaned, 
the oil for the bath must be rain water or condensed steam. On 
no account should hard water be used or water from a source 
suspected of containing acid, chemicals, or other oils. When the 
water has been poured in warm (130F.) the right quantity of 
oil can be added usually a layer from J^ to ^ in. thick, which 
equals from 3 to 6 per cent of the volume of water in the bath. 

The engine is now run slowly at light or no load, until the bath 
gets well emulsified; not until then should the full load be put on, 
and only after examining the bath. For this purpose the engine is 
stopped, and one of the doors removed; the oil will now be seen 
covering the surface; and after the surface is stirred with a stick, 
the water underneath must appear milky, yellowish white. If 
after being stirred the oil flows quickly together in a thick film, 
there is too much of it in the bath ; if it closes over the water only 
with difficulty, more oil must be added. 

During operation of the engine more or less water from the 
cylinders (condensation), particularly with wet-steam conditions, 
finds its way into the crank chamber, and the level of the bath 
rises. An automatic overflow should therefore be fitted; other- 
wise the oil overflows through the end bearings, and the engine 
may run short of oil. Figure 83 shows such an overflow arrange- 
ment, which will be readily understood. When the level (1) 



rises, water from a quiet corner in the bath enters the inlet to the 
overflow pipe (2), and only the small amount of emulsified oil 
carried away with the water is lost. Any oil carried in suspension 
will be retained by the oil separator (3) and can be returned to 
the bath through the vent pipe fitted higher up on the crank 
chamber, together with the daily or weekly make-up for loss in oil. 
In the Westinghouse engines the make-up oil is fed through oil 
cups to the main bearings and, leaving these, reaches the bath. 

FIG. 83. Water-overflow arrangement. 

When superheated steam is used, and no condensation reaches 
the bath, some water will evaporate, and it may be necessary to 
add condensed water to the bath to keep up the level. In such 
cases the crankcase oil consumption with a good-quality oil 
becomes exceedingly low. Consumptions as low as 1 pt. per 
24 hr. for a 1,000-hp. Willans engine are on record. 

The greater the stream of water leaving the overflow (4) the 
greater the oil consumption; but under reasonably good condi- 
tions, a consumption of 0.5 to 3.0 pt. per 24 hr., according to the 
size of engine, will prove ample. 



There are, however, special sources of oil loss, such as loss 
through end bearings or past the pistons. Figure 84 shows the 
oil-thrower arrangement of a Willans engine. Any lubricant 
reaching the oil thrower (1) is returned to the bath through the 
passage shown. Leakage may occur if the thrower is too far 
away from the cover (2); drops of oil are ordinarily caught 
between the edge of the thrower and the beveled edge on the 
cover; but if the space is too great, oil may get past, without 
touching the thrower. When the oil has got past this point it 
may either leak down the outside of the cover or pass along the 

shaft. To prevent the latter trouble, the 
clearance between the shaft and the cover 
must be sufficient so that drops of oil may 
exist on one surface without touching the 

Leakage of oil may also be due to the 
overflow's being choked with emulsified 

FIG. 84. Willans oil 

FIG. 85. Willans stuffing 

clots of oil, cotton waste, etc. In that case the bath level rises, 
the oil finally overflows through the bearings, and, getting caught 
by the rim of the flywheel, is thrown into the engine room in 
the vicinity of the flywheel. 

With large shaft diameters the arrangement shown in Fig. 84 
is not always satisfactory, and a proper gland may be provided, 
with very soft packing, which must be tightened up very gently 
to prevent " grooving " of the shaft. To avoid such wear's taking 
place on the shaft itself, a bushing is provided as shown in Fig. 85. 

When the piston rings in a Westinghouse engine are in bad 
condition, the oil splash from the crank chamber will be drawn 


past the low-pressure pistons in particular and is exhausted 
with the steam. The quality of the oil has nothing to do with 
this trouble, and the only remedy is to put the rings in order; 
rounding the upper edges of the rings is always a good precaution, 
as on the upstroke the rings will ride on the oil film and on the 
downstroke will scrape off excess oil. 

Temperature. The presence of water in the bath ensures that 
the bearings shall not reach a temperature higher than 212F., but 
for normal running it is preferable to keep it much lower, say 
120 to 140F., after a 2- or 3-hr. run. 

With condensing engines this temperature is rarely exceeded, 
but with noncondensing engines the greater amount of heat 
from the cylinders often makes the bath uncomfortably hot; 
a simple arrangement of cooling pipes should then be fitted, and 
the boiler-feed water may be used as cooling water on its way to 
the feed pump. Approximately 25 per cent of the feed water 
will suffice to keep the bath reasonably cool. 

There should preferably be no joints in the cooling coils inside 
the crank chamber to avoid leakage, as, if the water is hard or 
contains acid, chemicals, or other impurities, a leakage into the 
bath may destroy the oil. For a similar reason cooling waters 
should be avoided which are liable to attack the cooling coils, 
as even pinholes will allow a great deal of water to leak in. 

Oils. If the bath were made with oil, only the bath tempera- 
ture would quickly rise, owing to the large amount of heat devel- 
oped by fluid friction in the bearings and by the splashing of the 
cranks through the viscous oil. When the bath contains only a 
small percentage of oil, the viscosity of the emulsion is practically 
the same as for water alone. This was shown by K. Beck of 
Leipzig. 1 The viscosity, taken by an Ostwald viscometer, of a 

10 per cent mixture of castor oil and water was only a little over 
1 per cent greater than the viscosity of pure water. It is rather 
curious that even with fairly high bearing pressures the water 
emulsion is capable of furnishing adequate lubrication. The 
explanation is probably that little particles of emulsified oil, or of 

011 in suspension, attach themselves to the rubbing surfaces and 
form a coating which prevents metallic contact; and the friction 
is very low because of the low viscosity of the emulsion, which 
forms the lubricating film. 

1 Zeitschrift fUr Physikalische Chemie, vol. LVIII. 


In the early days castor oil was the favorite lubricant, but it 
has several drawbacks; it becomes acid and gummy, owing to 
oxidation, as it is intimately mixed with the hot air in the crank 
chamber. While, therefore, castor oil produces excellent lubrica- 
tion and a rich emulsion, it often leads to corrosion of the surfaces, 
and a sticky deposit accumulates on the connecting rods, etc. 
The consumption of castor oil is comparatively large, as the water 
leaving through the overflow is heavily charged with emulsified 
castor oil. Castor oil is now seldom used; the same oil is used in 
the bath as in the steam cylinders and valves. 

In America straight-mineral dark cylinder oils are generally 
used ; and while they give a moderate degree of satisfaction under 
the best conditions, the results are not at all good when the 
engines employ wet steam or when certain hard limy boiler-feed 
waters are used. The emulsion, which is always poor with a 
straight mineral oil, breaks down under these latter conditions, 
resulting in high friction and wear. If the American users of the 
Westinghouse type of engine knew the results that are obtained 
by the use of lightly compounded filtered cylinder oils, the dark 
" Virginia' 7 and similar oils now used straight would soon be dis- 
placed by better oils. 

Dark cylinder oils, whether straight mineral or compounded, 
are inclined to become thick and livery, particularly if there is 
too much oil in the bath. In large engines, where the conditions 
are usually less trying than in small ones, dark compounded 
cylinder oils may, however, give good results; but in smaller 
engines, filtered cylinder oils suitably compounded with fixed oil 
are much to be preferred. When the steam is wet (boilers prim- 
ing), and boiler impurities are carried into the engine, some will 
reach the bath and will tend to thicken the oil. It is under these 
conditions that dark cylinder oils get "livery," whereas filtered 
cylinder oils are very little affected, even if there is rather too 
much oil in the bath. Filtered cylinder oils thus give a much 
greater margin of safety and prove more economical and more 
efficient than dark oils. It is a mistake to use a large percentage 
of fixed oil; 4 to 6 per cent is all that is required. With more 
fixed oil the emulsion becomes unnecessarily rich, and more oil 
is lost through the overflow. 

Small engines are sometimes lubricated with a bath of circula- 
tion oil, but 15 to 20 per cent of oil is then required as compared 



with 3 to 6 per cent when cylinder oil is employed. Certain small 
engines have the bearings more or less enclosed, and the oil holes 
are rather small. If cylinder oil were used in the bath, small clots 
of emulsified oil would choke these small openings, and circulation 
oils must therefore be used for such engines. 

Sticky deposits may develop on the rods, etc., as already men- 
tioned in reference to castor oil; similar black deposits may 
be produced with cylinder oils, particularly so with dark oils, 
and they will appear on the rods in peculiar patterns or streaks 
caused by the motion of the rods and consist of water, oil, oxi- 
dized oil (insoluble in petroleum spirit), and a small percentage of 
iron and iron oxide (wear). The cause of the deposits may be 
inferior mineral base in the oil (presence of too much coloring and 
bituminous matter) or inferior fixed oil (too much free fatty 
acid) ; or, again, the quality of the oil may not be at fault, but the 
temperature of the bath may be above 140F., which is a critical 
temperature as far as oxidation of the oil is concerned. Shortage 
of oil in the bath will also bring about deposits, but they will 
then be found rather rich in metallic contents, indicating exces- 
sive wear. 

For High-speed, Enclosed-type Engines, Employing the Splash-oiling System 


Grade of oil 


of oil used 

in bath 

Small, horizontal engines, operating 

in ordinary engine rooms. 
As employed in steam motor 

Vertical engines: 

Up to 50 hp 

Up to 300 hp. 
Above 300 hp . 

Circulation oil 1 or 2* 
Circulation oil 3 or simi- 
lar oil of even higher vis- 

( Circulation oil 2 or 
/ cylinder oil 2 F.L.C.f 
Cylinder oil 2 F.L.C. 
(Cylinder oil 3 F.M.C. or 
\ cylinder oil 3 D.M.C. 




4to 6 
4 to 6 
3 to 4 
3 to 4 

* For circulation oils, see p. 243. 

t For cylinder oils, see table page 408. 


In many modern high-speed engines, whether they be steam, 
gas, petrol, or Diesel engines, the crank chamber is filled with oil 
spray a more or less dense mist of fine oil particles. These 
engines are acknowledged to be safe and reliable in operation, as 
far as lubrication is concerned, notwithstanding what is probably 
a fact, that most of them when running would explode were a 
spark to be formed inside the crank chamber. 

It is a well-known fact that to make an explosive mixture 
with air, an inflammable gas of some kind is not essential. Any 
sufficiently inflammable substance in the form of fine dust will 
produce this effect if present in the requisite proportion, e.g., 
in the case of many explosions in coal mines. A mixture of air 
and coal dust can be made to explode when the coal dust reaches 
a certain percentage; and as soon as a spark or a naked flame 
is formed or brought within the danger zone an explosion will 
occur. An explosion of this character occurred in an oil-cake 
mill, the air being heavily laden with fine seed dust; the mixture 
was fired by a spark from a dynamo, and many lives were lost. 
Another explosion occurred in a flour mill, sparks from a hot 
bearing firing the mixture of air and flour dust. 

Coming back to the enclosed high-speed engines, it is obvious 
that, from the time of starting, an increasing amount of oil spray 
is formed owing to the smashing action of the moving parts on the 
stream of oil escaping from gudgeon pins or crossheads and 
crankpins. When the engine has been running for some time, 
the air will contain a certain constant amount of ''oil mist" in 
accordance with the conditions of speed, ventilation, etc., of that 
particular engine. In very large engines, e.g., large en closed-type 
marine Diesel engines, it is doubtful whether they ever contain 
sufficient oil mist to be capable of exploding; but in smaller and 
much higher speed engines, the danger of explosion is ever present. 

In 1911 an enclosed steam engine, 300 hp. with forced-feed 
circulation, exploded in a large hosiery factory. On a Monday 



morning the engineer had gone to the engine room, started the 
engine, and left the powerhouse; a few minutes later the engine 
exploded. This is what happened: 

During the week end the engineer had tightened up the brasses 
on the low-pressure crosshead, and on Monday morning the 
engine was started up without examination as to whether this 
bearing had been tightened up too much, which unfortunately 
was the case. After a few minutes the crosshead got hot ; the heat 
spread to the cast-iron slippers, which work vertically in circular 
guides about 8 in. in diameter. The clearance was only about 
0.01 in. when cold and 0.002 or 0.003 in. with the engine warm; 
consequently, the excessive heat conducted from the crosshead 
pin caused the slippers to expand and seize. The circular guides 
broke, and flying sparks from the slippers fired the mixture of air 
and atomized oil in the crank chamber. The governor casing 
blew off; the opposite wall of the engine room fell out, while one 
of the other walls was moved 4^ in. ; and the roof of the engine 
room was blown away. 

Another disaster took place on a British battleship. An 
enclosed steam engine exploded and killed a number of men. 
The papers reported that the explosion was due to carelessness 
on the part of one of the men, who approached the engine with 
a naked light just after it had been opened and the inspection 
doors removed, so that the crank chamber was still full of the 
mixture of atomized oil and air. 

Similar explosions have been reported in connection with 
Diesel engines installed in submarines belonging to one of the 
large Continental powers, and in several cases the explosion was 
due to sparks in the crank chamber owing to seizing of one of the 
pistons. This would seem to indicate that as far as cylinder lubri- 
cation is concerned, the greatest care should be taken in designing 
the lubricating system and in using such oils as will ensure as 
safe and clean lubrication as possible of the pistons, particularly 
so in the case of Diesel engines for naval purposes, where high 
speed and short connecting rods are the characteristic features, 
owing to the cramped space available for the engines. 

Several explosions have happened in the past with enclosed 
high-speed gas engines, due to exactly similar conditions, viz., 
a mixture of air and atomized oil. It is a fact worthy of note 
that at least one firm of engine builders in England now ventilate 


their enclosed gas engines and Diesel engines by fitting a small 
fan that removes from the crank chamber any gases that may 
pass the pistons, as well as the finest oil vapor, thus making the 
possibility of an explosion very remote indeed. 

This system appears to be particularly desirable for high- 
speed naval Diesel engines and has been used in Continental 
submarines; to avoid excessive loss of oil, the fan discharges 
through a separating tank, in which baffle plates cause a portion 
of the oil to " condense " and settle out. 



Hand oiling is still used, the practice being to pour oil from an 
oil feeder into oil cups, say during four to eight revolutions of 
the engine every half hour. The bearings get flooded with oil 
after each oiling, and thereafter the oil film is gradually squeezed 
out, and lubrication becomes less and less efficient until such time 
as the bearings are oiled again. Obviously, this method is both 
wasteful and inefficient. 

The better method now most frequently employed is to have 
oil cups fitted with siphon wicks which siphon the oil from the 
cups and deliver it into oil pipes leading to the various bearings. 
Siphon oil boxes are fitted near the tops of each cylinder and 
distribute the oil through feed pipes ending in "wipers," which 
are touched by oil-receiving boxes fixed on the moving parts, at 
the moment that these boxes reach their highest positions ; the oil 
is finally guided to the various points through pipes fixed to the 
moving parts. 

A siphon box is fitted over each main bearing with two or 
more oil feeds according to the size of the shaft; from these boxes 
may also be taken oil feeds for the crankpins, when the latter 
are arranged for "banjo" oiling. An oil box is fitted for each 
crosshead guide, and a comb fitted to the bottom end of the 
slipper dips into the oil well and carries the oil up on the guide 
which is usually water cooled. 

The oil feeds vary with changes in temperature, the oil feeding 
more quickly when warm, owing its lower viscosity. The oil 
feed is much dependent on the oil level in the cups. The siphons 
feed more slowly when the oil level is low; it is therefore neces- 
sary to keep it as uniform as possible by frequently replenishing 
the oil cups. 

A better system is to replenish the various siphon oil cups not 
by hand but from a centrally placed oil tank, feeding adjustable 
quantities of oil through the feed pipes, each of these having a 


sight-feed arrangement by which the oil feed can be ascertained 
going to the corresponding oil cup. If, for example, one feed 
pipe is feeding 60 drops per minute to one of the oil cups, the 
latter distributing by siphons the oil to several points, then the 
oil level in this cup will quickly adjust itself automatically to 
such a level that the oil siphons, all told, will siphon out 60 drops 
per minute. If they feed more, the oil level will gradually 
decrease until a point is reached when the oil feeds, all told, 
amount to 60 drops per minute. The control of the oil feeds from 
the central oil tank can best be done by mechanically operated 
lubricators, which start and stop feeding with the engine. 

Experience has proved that the installation of such a central 
distributing-oil tank, preferably in connection with mechanically 
operated lubricator pumps, will save from 40 to 60 per cent of the 
total amount of oil consumed for external lubrication. 

There is always a greater or lesser amount of condensed steam 
finding its way down the piston and valve rods and dropping all 
over the external moving parts, and in case of a hot bearing the 
cold-water hose is frequently applied. Sometimes a small trickle 
of water is allowed to run into or on to those bearings which are 
inclined to run rather warm. When oils pure mineral in charac- 
ter are used, the water will displace the mineral oil, and the bear- 
ings will heat and may seize. 

Marine-engine oils should therefore be compounded with a 
suitable percentage of good-quality fixed oil, so that they will 
emulsify freely with water and form a rich and creamy lather. 
Good-quality marine oils, while they combine satisfactorily with 
water, will give more efficient and more economical lubrication 
if they are used without water. The oil when leaving the bear- 
ings, usually in a more or less emulsified condition, is run to waste 
into the bilges, as it is impossible to recover it from the emulsified 
waste oil. 

Marine-engine oils should contain only a small percentage of 
fatty acid, say less than 2.8 per cent F.F.A., so as not to cause 
corrosion or pitting. The fixed oil should not produce a dis- 
agreeable odor exposed to the heat in the engine room, nor 
should that used for compounding be of a drying nature, but 
semidrying oils like rape oil will give good service. Castor oil 
was at one time much used and still is, largely in the East, but 
is expensive when used alone. It can, however, be mixed with 


mineral oil in the presence of an animal oil, say 20 per cent 
castor, 6 per cent lard oil, and 74 per cent heavy- viscosity mineral 
oil preferably Texas, Russian, or other asphaltic base to get a 
low setting point. The lower the setting point and the greater 
the percentage of good-quality fixed oil of reasonably low cold 
test the less will the oil be affected by climatic changes. 

The bearings of large marine steam engines require oils of 
great oiliness to give the necessary margin of safety under the 
severe operating conditions. Pure mineral oils of the requisite 
oiliness do exist, but they are so viscous that they will not siphon 
properly, and they feed irregularly owing to poor cold tests. The 
admixture of fixed oils having great oiliness is therefore dictated 
not only by the presence of water but also by the necessity of 
keeping the cold test and viscosity of the finished oil reasonably 
low. Blown rape oil, blown cod oil, or blown whale oil, prefer- 
ably the first, are used to the extent of 10 to 25 per cent, the fixed 
oils being usually blown until they have a viscosity of 720 to 
1400 sec. Saybolt at 210F. Such very viscous fixed oils 
raise the oiliness and the viscosity of the mineral base appreciably 
without unduly raising the setting point. 

The table below gives typical readings of three marine-engine 
oils which will serve for all marine purposes where compounded 
oils are needed. The important figures are those for viscosities, 
cold tests, and compound; the figures for specific gravity and 
flash point are of little consequence. 

Large engines and vessels navigating in hot climates need 
higher viscosity oils than smaller engines or vessels operating in 
colder climates, but only practical tests extending over a period 
of, say, 3 months can decide which of the three grades should 
be preferred and what percentage of compound it should contain. 





poises at 



per cent 







'lO to 20 






2 %o 

10 to 20 






3 %5 

15 to 25 

* See table, p. 57. 


The high-viscosity oils are usually the more economical, but 
they may be unnecessarily viscous and thus waste power in 
creating too much "oil drag" in the bearings. 

Forced circulation has during recent years made great progress 
in naval ships both in Europe and in the United States, not only 
for steam turbines and auxiliary high-speed enclosed-type steam 
engines but also for the large reciprocating-type main engines 
in destroyers and other craft. 

The working parts including the crossheads are enclosed in an 
oiltight casing, packed glands being provided for the piston and 
valve rods to prevent too much water from getting into the oil in 
circulation. Observation windows and electric lights may be 
fitted to watch the moving parts. The oil is supplied by recipro- 
cating pumps operated by the engine itself or independent 
thereof. It may be forced into the hollow crankshaft, holes 
being drilled radially at each main bearing, crankpin, arid eccen- 
tric, the oil from the crankpin continuing its way through a 
tube fitted to the connecting rod into the crosshead bearing, 
finally reaching the crosshead guides. 

The oil-delivery pipes may also deliver the oil into the main 
bearings first of all, the oil thence reaching the hollow crankshaft, 
etc., as is customary on land engines. 

The oil-supply system is frequently made in duplicate. 

The oil collecting in the oil wells is pumped away by indepen- 
dently operated pumps, fitted with suction strainers, being 
delivered through a filter to a "settling tank," finally reaching 
the storage tanks ready to be circulated afresh. 

Main engines fitted with forced oil circulation have been 
inspected after the vessel has done 20,000 miles; the toolmarks 
in the white-metal bearings were found to be still visible, and no 
measurable wear had taken place. 

The risks of accidents due to hot bearings caused by insuffi- 
cient oil supply are practically eliminated by this system; there 
is a great saving in the time and expense that were required 
with siphon lubrication in rebabbitting, adjusting, and examining 
bearings after each voyage. The cost of lubrication is also 
much reduced by forced oil circulation, and the engines operate 
much more quietly. 

Auxiliary engines which are now frequently fitted with a 
full force-feed circulation system should preferably have the 


cylinders raised so high above the crank-chamber top that no 
part of the piston and valve rods entering the cylinder or valve 
glands will enter the scraper glands fitted in the crank-chamber 
top. If this is arranged, no oil from the crank chamber can 
possibly enter the steam cylinders, which is important with a 
view to preventing oil from reaching the boilers. 

The oils used for force-feed circulation systems must be similar 
to those used for steam turbines; i.e., they are circulation oils (see 
page 243). Circulation oil 2 is used on most small auxiliary high- 
speed engines; circulation oil 3, on large slower speed engines and 
larger auxiliary high-speed engines. 

One might ask why those oils pure mineral in character and 
lower in viscosity than the compounded marine-engines oils can 

FIG. 8$. Horseshoe thrust bearing. 

replace the latter and with such great success. The answer is 
simply that the oil is supplied to all bearings in abundance, not 
only supplying a complete lubricating film but also continuously 
removing frictional heat from the bearings, which therefore run 
much cooler. The bearings do not become contaminated with 
water, and as the revolving parts practically "float" on a com- 
plete oil film, wear is almost eliminated, and the results are excel- 
lent from every point of view. 

It is to be hoped that the merchant marine will take advantage 
of the experience gained by the various navies with force-feed 
circulation, which undoubtedly is very superior to the systems 
now generally employed. 

Thrust Bearings. Figure 86 illustrates the horseshoe type of 
thrust bearing still almost universally used for marine steam 
engines. The collars on the shaft press against the horseshoes, 
which are adjustable in a fore-and-aft direction so as to distribute 
the load more or less evenly between them. Changes in tem- 
perature difference between shaft and horseshoes alter the distri- 



button of load and cause heating of certain collars and shoes, 
so that a hot thrust is by no means uncommon; in fact, the thrust 
bearing often gives the engineers more trouble than any other 
bearings or part of the engine-room machinery. The oil is fed 
by siphons from the top and led into oil grooves of various 
" fancy" patterns. The collars are often arranged to dip into 
the oil and carry it up with them. The oil bath is sometimes 
fitted with a cooling coil, but it is more effective to cool the 
horseshoes themselves, which is often done in large and important 
thrust bearings (see Fig. 87). 

The lubrication is, however, always poor, as the centrifugal 
force throws the oil away from the points where it is most needed ; 

FIG. 87. Cooling the thrust. 

the bearing pressures allowed are therefore low usually 50 to 
70 Ib. per square inch with a mean surface speed of 500 ft. per 
minute, but with the best cooling arrangements and perfect 
workmanship a bearing pressure of 100 Ib. per square inch and a 
mean surface speed of 600 ft. per minute has been accomplished, 
and in other cases a pressure of 60 Ib. per square inch with a 
surface speed of 800 ft. per minute. The friction is high, often 
consuming 5 per cent of the shaft horsepower; the rubbing sur- 
faces are in only a boundary-lubricated condition, the coefficient 
of friction being approximately 0.03. 

The Michell single-collar thrust bearing (see page 175) will 
no doubt come more and more into general use, not only for 
marine turbines but also for marine steam engines, as with a 


little intelligent attention it gives no trouble whatsoever and 
consumes less than one-tenth of the friction ordinarily wasted 
in the horseshoe type of thrust bearing. 

Stern-tube Lubrication. A large majority of ships are fitted 
with lignum-vitae stern-tube bearings, the propeller shaft being 
fitted with a bronze liner, and the lignum vitae being placed as 
strips 2 to 3 in. wide with 2-in. spaces between the strips. Salt 
water is usually the only lubricant used in these bearings, but 
occasionally the stern-tube gland is fed by a Stauffer grease 
cup through which a suitable grease can be fed into the gland 
with a view to reducing the considerable amount of friction 
generated here. 

Unquestionably, there is a very great frictional loss in the 
lignum-vitse stern-tube bearings, and the only way to reduce 
this frictional loss is to fit the bearing with an outer gland, as in 
the case of the Cederwall, Vickers, or similar type of packing. 
Thus enclosed, the lignum vitae can, if desired, be replaced by 
proper bearing metal, and in any case the stern-tube bearing can 
be efficiently lubricated by means of oil or thin grease. This 
means a great saving in power and also entails the advantage 
that where the vessel gets into shallow waters, as is the case with 
a number of river boats or coasting steamers, the entrance of mud, 
sand, or other impurities is entirely obviated, thus preventing 
trouble and giving much longer life to the stern-tube bearing, 
the wear being practically eliminated. 

Another advantage is that galvanic corrosion, rusting, and 
pitting of the shaft cannot take place, assuming, of course, that 
the lubricant employed is of reasonably good quality. 

An arrangement patented by Vickers and Sons, Leeds, is 
illustrated in Fig. 88 and was referred to in Engineering. They 
write as follows: 

This appliance was fitted to two twin-screw hopper barges con- 
structed for the Clyde Navigation Trustees by Messrs. Fleming and 
Ferguson, Limited, of Paisley. After running two years the shafts 
were examined and were found to be in very good condition. They 
were again examined after 3 years' continuous work, and the 
wear was found to be less than ^ 2 in. of the total diameter of the 
shaft in the bush, so that it was not considered necessary to true 
up the bushes. When one remembers the peculiar gritty nature of 
the Clyde water and that the barges are often in close proximity to 


little intelligent attention it gives no trouble whatsoever and 
consumes less than one-tenth of the friction ordinarily wasted 
in the horseshoe type of thrust bearing. 

Stern-tube Lubrication. A large majority of ships are fitted 
with lignum-vitse stern-tube bearings, the propeller shaft being 
fitted with a bronze liner, and the lignum vitae being placed as 
strips 2 to 3 in. wide with 2-in. spaces between the strips. Salt 
water is usually the only lubricant used in these bearings, but 
occasionally the stern-tube gland is fed by a Stauffer grease 
cup through which a suitable grease can be fed into the gland 
with a view to reducing the considerable amount of friction 
generated here. 

Unquestionably, there is a very great frictional loss in the 
lignum-vitse stern-tube bearings, and the only way to reduce 
this frictional loss is to fit the bearing with an outer gland, as in 
the case of the Cederwall, Vickers, or similar type of packing. 
Thus enclosed, the lignum vitae can, if desired, be replaced by 
proper bearing metal, and in any case the stern-tube bearing can 
be efficiently lubricated by means of oil or thin grease. This 
means a great saving in power and also entails the advantage 
that where the vessel gets into shallow waters, as is the case with 
a number of river boats or coasting steamers, the entrance of mud, 
sand, or other impurities is entirely obviated, thus preventing 
trouble and giving much longer life to the stern-tube bearing, 
the wear being practically eliminated. 

Another advantage is that galvanic corrosion, rusting, and 
pitting of the shaft cannot take place, assuming, of course, that 
the lubricant employed is of reasonably good quality. 

An arrangement patented by Vickers and Sons, Leeds, is 
illustrated in Fig. 88 and was referred to in Engineering. They 
write as follows: 

This appliance was fitted to two twin-screw hopper barges con- 
structed for the Clyde Navigation Trustees by Messrs. Fleming and 
Ferguson, Limited, of Paisley. After running two years the shafts 
were examined and were found to be in very good condition. They 
were again examined after 3 years' continuous work, and the 
wear was found to be less than H2 in. of the total diameter of the 
shaft in the bush, so that it was not considered necessary to true 
up the bushes. When one remembers the peculiar gritty nature of 
the Clyde water and that the barges are often in close proximity to 



dredges which are disturbing the bed of the river, this result will be 
accepted as very satisfactory. The section is almost self-explanatory. 
It will be seen that on each side of the floating packing there are two 
packings; and next to the guard ring there are elastic disks which grasp 
the shaft like the sleeve of a diver's jacket. The inner one is a fine 
elastic woollen felt, and the outer of a special composition of a slightly 
elastic nature which is unaffected by either sea water or oil. Inci- 

1 Floating Packing! 

2 Elaitic DUc Packing 

3 White Metalled Bushing 

4 Oil Supply from Tank abore 

5 Oil Feed Pipe 

6 Oil Overflow Pipe 

7 Handpump Oil Supply 

8 Gland with Soft Packing 

FIG. 88. Vickers' stern-tube packing. 

dentally, the application of oil here reduces the friction, and as the 
friction resistance within the stern tube is a large proportion of the total 
friction of the engine and shaft, the advantage is very considerable. 

The continuous bronze liners now often fitted, which are 
carried right into the propeller boss, protect the shaft from 
galvanic corrosion but do not prevent the entry of sand, so that 
whatever system of lining or bushing (cast iron, white metal, or 
lignum vitse) is employed, many advantages are always obtained 
by enclosing the stern tube and having a proper oiling arrange- 
ment fitted. 



Axle boxes. Axle boxes are termed inside or outside according 
to whether they are inside or outside the wheels. Generally 
speaking, tenders and cars have outside axle boxes, and locomo- 
tives inside boxes; some locomotives, however, have the wheels 
inside the frames and the axle boxes outside. 

Figure 89 shows an outside axle box. A door is formed in the 
front portion of the box. To prevent rain water from entering 
the box through the joint, the box may project above the door, as 

FIG. 89, Outside axle box. 

shown; another solution is to have attached to the door a sheet- 
metal rain guard which projects over the top of the box (Fig. 90). 
For the same reason the door should be so designed as to prevent 
water from getting in at the sides and bottom. At the wheel side 
of the box is a dust guard, usually made of wood, in two halves, 
which are forced gently against the shaft by springs. One type 




of dust guard made of lignum vitse has oil pads fitted in little 
recesses in both halves; the bottom pad has two siphons, the 
ends of which are immersed in the oil reservoir and thus lubricate 
the dust guard and prevent wear. 

Most dust guards get little or no lubrication, and when they 
are worn they no longer keep the dust out so efficiently as one 
might desire. 

Between the top of the "brass" and the cover of the axle box, 
to which the weight is transmitted through the springs, is placed 
a hardened cast-steel liner or wedge piece, which serves to dis- 
tribute the load uniformly over the whole of the brass. 

FIG. 90. Axle box with rain guard. 

Inside axle boxes consist of two almost semicircular castings 
with vertical side plates which fit the horn plates; the lower half 
is suspended from the upper half by bolts, and the springs rest 
upon the upper half. 

Journals and Bearings. It has become a general practice to 
roll the journals of crankpins and axle journals with a hard-steel 
roller, in order to compress the surface and make it very tough 
and capable of resisting wear. The roller is held in the tool post 
of the lathe after the finishing cut has been taken and is forced 
against the journal. This same method is also frequently used 
for rolling the white metal in babbitted bearings. 

As regards bearing metals, locomotive driving and trailing 
bearings are usually bronze lined with white metal, and the tend- 
ency is to extend the use of white metal as a lining for bearings. 
The reason for this is that a good white metal combines the 


necessary strength with plasticity. It contains hard grains 
which transmit the pressure to a plastic matrix. The hard grains 
prevent excessive wear, and as they are embedded in a yielding 
matrix the load is evenly distributed over the entire surface. 

With phosphor bronze, unless the bearings are very carefully 
scraped together, the load is not so evenly distributed; and in the 
case of shocks and vibration, local heating may easily occur, 
causing a hot bearing. 

It is a well-known fact that in running down a long gradient, 
crankpins with bronze bearings are liable to heat, owing to exces- 
sive shocks in the bearings caused by the absence of steam in the 
cylinders, which otherwise would " cushion" the blow at either 
end. Strips of white metal embedded in the crankpin bearings 
help to prevent such heating. 

Another reason for the wider adoption of white metal is that 
should the bearing seize, the shaft is only little affected, and the 
bearing can be rebabbitted at a small cost. 

A large proportion of lead in white metal is not desirable, as it 
causes increased friction and, being a bad conductor of heat, 
does not allow the heat to be dissipated so readily; consequently, 
the bearings run warmer. Furthermore, lead is more easily 
attacked by acids which may be present in the oil. 

It is necessary for the white metal to be supported by brass or 
cast iron of sufficient thickness to avoid distortion under running 
conditions. If the brasses are too light, they may crack or at 
least run exceedingly warm. This action causes the edges of 
the brass to pinch the journal and makes it very difficult for the 
oil to do its work properly. 

As mentioned above, phosphor bronze can be used as a bearing 
metal only when the faces are very accurately scraped together. 
In the case of white metals, however, such careful fitting is not 
necessary, as the bearing surfaces will bed themselves together 
more readily. 

Of recent years, bronzes of a new type called " plastic bronzes" 
have been used, particularly in the United States. The differ- 
ence between them and the white metals is that they are made 
up of plastic substances embedded in a hard matrix, whereas 
the white metals are made up of hard substances embedded in a 
soft matrix. There seems to be a divergence of opinion as to 
the utility of these plastic bronzes. 




It is very important that the load on journals shall not be trans- 
mitted eccentrically. Take a journal with a diameter D and 
length L, the load being not in the center but transmitted at a 
point x in. away from the center (Fig. 91); then the bearing pres- 
sure at the extreme ends of the bearing will be 


To take an example: Let P = 16,000 lb., D = 8 in., and L = 
10 in. If the load is central, the pressure per square inch will be 

200 lb. uniformly distributed. If 2 in. 
are added to the inside of the box, 
making the length 12 in. and x equal 
to 1 in., then the maximum and min- 
imum pressure per square inch will be 
249 and 83 lb., respectively, at the 
outside and inside edges, while the 
average pressure is only 166 lb. per 
square inch. This will indicate that 
it is often preferable to accept an 
I increased pressure per square inch 

I rather than create an eccentric load- 

FIG. 91. Eccentric loading. 


On locomotive driving journals the 
brass covers half the journal, and the pressures per square inch 
are usually somewhere about 200 lb. 

In the case of car and tender bearings, the arc over which the 
brass touches the journal is usually 90, occasionally 120 deg., and 
the pressure per square inch of projected area is usually from 
300 to 325 lb. 

The small space available, particularly on narrow-gauge rail- 
ways, often makes it difficult to give locomotive bearings the 
dimensions required for cool running. The journals can always 
be made strong enough, but the difficulty is to make them long 
enough. When a bearing runs consistently hot, an increase in 
journal diameter is no remedy, as, although the bearing pressure 
per square inch is reduced, the surface area and surface speed of 
the journal are both increased, so that, notwithstanding the 



larger radiating surface, no advantage is obtained. With greater 
length of the journal, the surface area is increased, but not the 
surface speed, and the result is a cooler running bearing. 

Some interesting information was given by Robson in an 
article 1 in which he gives an empirical formula for judging whether 
a bearing will be inclined to overheat or not. 

Let S = the maximum continuous speed of the vehicle in miles 

per hour. 

D = the diameter of the wheel in inches. 
W = the weight on journal in tons. 
L = the effective length of the journal in inches. 

K = 


K being a constant, which is determined by actual experience. 

The article gives values for this constant for different bearings, 
all of which are white metaled and, except in the case of crank- 
pins, lubricated by means of a pad or oil-saturated waste below 
the journal. 

Robson gives his experience with various bearings inclined 
to heat and with others that, owing to longer journals, ran reason- 
ably cool. 

A summary of his recommendations is given in the following 
table : 


Type of bearings 

miles per 

Value f or K 


Inside locomotive journals on carrying axles and 




Outside journals on locomotives and tenders . . 



Crankpin journals 


4 to 4 5 

(For some inside crankpins K was 5.6 which was 

too high but could not be reduced on account of 

the narrow track) 

Carriage journals 



Goods and mineral-wagon journals 



Engineering, Nov. 25, 1910. 



Locomotive Axle Boxes. The usual practice is by means of 
siphon oil feeds (tail trimmings) from auxiliary oil boxes, the 
oil being led through tubes to the top of the bearings, entering 
the bearing through either a central oil hole into one longi- 
tudinal oil channel at the top of the brass or two oil holes leading 
into two oil grooves forming a slight angle with the journal. By 
this system the oil enters the bearing only with difficulty, except 
at the two bearing ends, and, once it has left the bearing, is lost. 

In modern systems the boxes are fitted with oiling pads 
underneath the journals, or they are filled with waste, preferably 
woolen waste thoroughly saturated with oil. The oil that enters 
the bearing is caught by the pad or the waste and distributed 
over the entire underside of the journal. The lower edges of 
the brass are eased away, so as to facilitate the entrance of the oil 
film between the journal and the brass. 

The most recent practice is to install mechanically operated 
forced-feed lubricators on the frame or in the cab, from which 
the oil is automatically distributed to the axle boxes under pres- 
sure. Test cocks are provided in suitable positions, so as to . 
regulate and test the oil feed. This method is an ideal one, as it 
ensures a feed of oil to the bearings in direct proportion to the 
revolutions of the journal; also, it is unquestionably the most 
economical, and the oil reaches the bearings with absolute cer- 
tainty, the distribution being entirely automatic. 

Where mechanical lubricators are used for feeding oil both 
to the cylinders and to the axle boxes, such lubricators should 
have two compartments, so that a bearing oil may be used for 
the axles, and cylinder oil for the cylinders. Obviously, it is 
ordinarily not desirable to use cylinder oil for the axle boxes, as 
it is far too viscous and causes unnecessarily high temperatures 
of the journals and boxes. 

In the case of bogie boxes, oiled through siphons from the top, 
they are exposed to rain or to the spray of water from the cylinder 
waste-water cocks. If sufficient water enters the oil well on top 
of the box, it will dislodge the oil and thus cause a heated journal. 
There is a general tendency among engine drivers to fill up the 
oil wells too high, and during running the vibration and oscilla- 
tion cause the oil to splash over the edge of the box, causing 



unnecessary waste. To overcome this, the best method is to 
fill the oil well with saturated waste, interlacing the oil siphons 
into it, and oil can then be added to the waste as required. 
This will prevent the entrance of water and will also prevent 
waste of oil. 

Axle Boxes for Tenders and Cars. In many cases pads 
are used for the underside of the journal, plus an additional oil 
feed by means of siphons arranged in the top of the boxes. The 
best practice is to use a pad or waste in the boxes and rely on 
these for the lubrication without any additional oiling from 
above; this permits doing away entirely with oil grooves in the 

FIG. 92. Pad oilers. 

bearings, so that the whole bearing surface is available to carry 
the load. 

Pad Oilers. The best known make of these oilers is the Arm- 
strong (Figs. 89 and 92), which has given general satisfaction 
and is extensively used. The Armstrong oiler consists of a pad on 
a light frame, supported by resilient steel springs. The pad is 
so woven that the points of the pile only lightly touch the journal. 
This pile is made of a special mixture of cotton and wool in order 
to retain the oil drawn up from the well of the box by the feeders, 
which should have high capillary powers. The buttons, which 
are made of lignum vitae, act as buffers and prevent the pile of 
the pad from being flattened out and glazed; in this way the 
capacity of the pad for supplying oil to the face of the journal 
remains unimpaired for a long period. New oilers should be 



dried and soaked in oil for about 12 hr. before being placed in the 
axle boxes. About 1 pt. of oil should be supplied to each axle box, 
or sufficient to cover the bottom of the well to the depth of % in., 
and a similar quantity about every 3,000 miles. If the axle boxes 
are dustproof, and the oilers are kept free from grit and properly 
fitted, the makers claim that they will last 250,000 miles without 
repair or removal and guarantee that they will last for 100,000 

Pad oilers like the Armstrong will lubricate the journal how- 
ever high the speed may be, and the action is unaffected by 
frequent changes in the direction of rotation. 

The use of such oilers results in : 

Ample and uniform oil distribution. 

Freedom from hotboxes under most conditions. 

Less necessity for frequent periodical inspection of axle boxes. 

r' ( 2& 'Section 
FIG. 93. Waste oiling. 

Reduction in oil consumption and other general lubricating 

Waste Oiling. Good wool waste should be soaked with the 
proper seasonable kind of oil far at least 48 hr. before being used. 
The surplus oil should be drained off, allowing sufficient oil to 
remain so that it will show under slight pressure. If there is too 
much oil in the waste, the latter becomes too heavy and will fall 
away from the journal, thus depriving the bearing of lubrication 
altogether. Well-soaked waste will have absorbed approxi- 
mately five times its own weight of oil. 

The first waste (Fig. 93, A) should be moderately dry and 
packed tightly around the back end of the box, so as to make 
a guard for the purpose not only of retaining the oil but of exclud- 
ing the dust. Then the box should be packed with the drained 



waste, made into balls, firmly enough so that it will not fall away 
from the journal when the car runs over crossings, etc., but not so 
tightly as to squeeze out the oil. The waste should be kept even 
with the journal, an inch below the edges of the brass. This is 
most important, as waste packed too high will be caught and 
carried round, causing a hotbox. 

At high journal speeds, say about 300 ft. per minute, the waste 
is inclined to be pushed over to one side of the box by the friction 


Steel Handle 


Hook to Open and Handle 
Box Lidi 

FIG. 94. Packing tools. 

between the journal and the waste and there compressed so tightly 
that lubrication becomes deficient. There is one type of box 
that has three compartments divided by longitudinal ribs, 
thus effectively preventing the waste from moving and ensuring 
its uniform saturation all through. 

The waste in the front end of the box should be as high as the 
opening and have no thread connection with that underneath 
the journal. This waste should be placed in the box by hand 
after the box has been packed. It performs no service other than 


to act as a stopper to prevent the waste that is doing the work 
of lubrication from working forward. 

It is important to give some intelligent attention to the waste 
in the boxes during service, the chief requirement apart from oil- 
ing being lightly to loosen the waste packing on either side of the 
journal for about every 1,000 miles' run, to bring it into good 
contact with the journal and avoid the hardened and glazed 
condition which is gradually brought about by contact with the 
revolving journal. Suitable tools for this purpose and also for 
packing the boxes are illustrated in Figs. 94 A, B, and C, showing 
a packing knife, hook, and loosening tool, respectively. 1 

Dust Guards. Efficient dust guards to prevent the entrance 
of dust are of the very greatest importance. Too much attention 
cannot be paid to this matter, as, if dust and grit are allowed to 
enter, the lubrication can never be perfect, and pad oilers and 
waste are liable to be choked. The dust trouble is particularly 
prominent in countries like the south of England, owing to the 
lime dust. 

In the case of newly laid roads, it frequently happens that fine 
granite dust causes trouble; being very hard and fine it enters the 
boxes and may cause a great deal of wear. 

Inspection and Oiling of Axle Boxes. Although, as a general 
rule, it is true that regular and careful inspection of axle boxes is 
desirable, yet it is also true that there can be too much inspection. 
As a matter of fact, pad oilers (and this also refers to woolen 
waste), once they are well fitted and work well, should not be 
disturbed in any way. An examination every 3 months will, 
as a rule, be quite sufficient, and at the same time a small quan- 
tity of oil may be introduced in the box, assuming that there is no 
additional oil supply from the top. 

The oil consumption with waste packing ranges from 500 to 
4,000 miles per pint of oil, a good average being 3,000 miles per 
pint of car oil. 

Special Oiling Systems. Lubrication of axle boxes by means 
of a circulation system has attracted considerable attention. 
Several systems have been tried, including a force-feed circula- 
tion system by means of a rotary oil pump and also a system con- 
sisting of a round disk fixed to the front end of the journal, 
dipping in the oil in the bottom of the box and lifting it to the 

1 Copied from American Locomotive Dictionary. 


top of the box, from which it flows into the bearing in liberal 

It is obviously desirable (particularly in railway practice) to 
give the journals as liberal a supply of oil as possible. The 
difficulties are that it is not easy to prevent excessive leakage of 
oil through the ends of the box and that the entrance of dust and 
dirt makes the oil dirty and may cause clogging of oil pipes where 
such exist. Other mechanical appliances have been tried, such 
as rollers against the underside of the journals, but have not been 

It must be kept in mind that whatever appliance is used, it 
should be so designed that it is not liable to get out of order; e.g., 
the clogging of an oil pipe or the breakage of one due to vibration 
will cause stoppage of the oil supply altogether, with disastrous 

During late years, considerable progress has been made in 
employing ball and roller bearings for axle boxes; the lubrication 
of such bearings represents much fewer problems than is the case 
with axle boxes for ordinary journal bearings. 

Connecting Rods and Other Parts. The brasses in connecting- 
rod bearings must be let completely together so as to cover the 
entire surface of the journals and minimize the entrance of dust 
and grit. 

Siphon lubrication is extensively used. For those parts which 
require only a small amount of oil, trimming pins or trimming 
plates are used, being a piece of ^-in. wire or Hs-in. plate, which 
has a hole at the bottom and also at the top, through which are 
threaded one or two strands of wool, just sufficient for proper 
lubrication of the motion bars or other parts, where such a small 
amount of oil is found ample. 

It is safer to use siphons than to feed the oil through oil cups 
where the oil feed is adjusted by means of a needle valve, as a 
needle valve is more easily choked than a siphon. Figures 95, 
96, and 97 show various designs of such oilers. 

For reciprocating parts, such as connecting-rod ends, choke or 
plug trimmings are frequently used (see Fig. 98). This trimming 
is pushed well into the siphon tube and prevented from dropping 
right into the tube by a big loop, which rests on the top of the 
siphon tube. When the engine is running, the oil is thrown up 
into the siphon tube, and, the trimming being, say, ^6 i n - below 



the top, a well or reservoir of oil is always maintained, the oil 
soaking through the plug trimming and entering the bearing. 
The plug trimming should preferably end close to the journal, as 

FIG. 95. FIG. 96. 

Locomotive stationary oilers. 

FIG. 97. 

this largely prevents the oil from being wasted by escaping 
between the brass and the strap. Sometimes a little tube is 
screwed in here, so as positively to prevent escape of oil. Plug 

FIG. 98. Plug (choke) trimming. FIG. 99. Rod needle oiler. 

trimmings may be made of copper trimming wire, the wire being 
wound in the same manner as the yarn in the usual plug trimming. 
The advantage is said to be that a much heavier oil can be used 


than could possibly siphon through the ordinary worsted trim- 
ming. Sometimes (in Continental practice) oil is allowed to go 
direct into the siphon tube through holes at the bottom; this, of 
course, means waste of oil while the engine is standing. In 
America and on the Continent, plug trimmings are frequently 
discarded in favor of small needle valves, consisting of a loose- 
fitting pin with a head at the top (Fig. 99), the upward and down- 
ward motion of the pin being regulated by an adjustable stop in 
the oil-cup cover. 

Another method is to have simply a long thin piece of wire bent 
over at the top, fitted in the siphon tube, passing through a small 
fitting screwed into the top of the siphon tube, and having a cen- 
tral opening through which the wire or needle passes down. The 
difference in diameter between the needle and the opening deter- 
mines the oil feed. 

In oil cups that are entirely enclosed, the cover should have a 
tiny hole to allow the air to get in as the oil leaves the cup, or 
the hole in the cover should be plugged up with a piece of cane 
(which is porous) or a piece of cork with a V groove at the side. 

When changing over from an oil largely vegetable or animal 
in character, it nearly always happens that the siphons and trim- 
mings get more or less choked with deposit due to the change. 
It is therefore to be recommended, wherever any drastic change 
in oils is to be carried out, that new trimmings be made for all 
the oil cups and lubricators. 

The consumption of engine oil for the various external parts of 
a locomotive, including axle boxes, varies considerably according 
to the size and the method of lubrication. The consumption 
may be as low as 2 and as high as 8, the average being about 

3H Pt- P er 10 miles - 

When sharp curves are frequent it is desirable to oil the wheel 

flanges by means of a jet of oily steam. Various forms of lubri- 
cators are employed for this purpose; they all endeavor to 
atomize the oil with a jet of steam, which is then directed on 
to the wheel flange. 

Methodical Oiling. It is very important that the oiling of the 
locomotive be carried out in a methodical manner, the oiler 
going round from one part of the engine to another, oiling always 
in the same manner of rotation. This is the only way in which 
he can be reasonably sure of not forgetting some of the parts. 



As a matter of fact, lack of attention to this point may be 
said to be very largely responsible for bearing troubles. This 
also applies to the attention that should always be given to 
taking out siphons or trimmings wherever possible when the 
locomotive has finished the journey. Overfilling of oil holes or 
oil cups should be avoided, as it is wasteful and does not improve 


In the United States the use of grease on locomotives has dur- 
ing recent years been given some considerable attention, not 

FIG. 100. Rod grease cup. 

only for the connecting rods and coupling rods but also for the 
axle boxes. 

Figure 100 shows the grease-cup arrangement for one of the 
rods; when the lock nut (1) is loosened, the threaded plug (2) can 
be given a turn and again locked; the grease gets squeezed into 
the bearing and is gradually consumed, until the plug is given 
another turn, and so on. 

Figure 101 shows the application of grease to a driving box; the 
grease is molded to the shape of the collar and placed on the 




000000000000 OjO^J*-*- 

g r ,*j". ? ? .... . ; rTr. . 

follower plate (1) ; the spring (2) pushes the follower plate upward, 
thus squeezing the grease through the perforated plate (3) shaped 
to the contour of the journal and kept at a distance of about 

Oil grooves are cut to distribute the grease as shown; the 
vertical grooves are cut only on the "off" side of the brass, 
presumably to act as drainage grooves. Through a hole shown 
on the left, some grease reaches the hub face of the wheel; a 
similar hole, not shown, is arranged for 
lubricating the horn plates. 

It is stated by the makers of these grease 
appliances that the grease recommended 
for the axle boxes must not get sticky when 
worked between the fingers and that when 
smeared with a penknife on a piece of white 
paper small bubbles of water must appear 
on the surface. The author has no per- 
sonal experience with these greases; they 
are probably rather soft low-melting-point 
greases somewhat similar to the English 
railway-wagon greases mentioned below 

and containing a certain amount of water FlQ - 101. Driving-box 

. grease lubrication. 

to bring about emulsmcation, so that the 

journal when revolving may continue to abrade or melt the 


It is obvious that whatever claims may be substantiated in 
the way of "economy" and ability to stand up to severe condi- 
tions, the amount of power lost in friction is considerably 
increased with grease lubrication and also the wear. 

An advantage with grease lubrication is that the starting fric- 
tion is lower than with oil on account of the thicker film between 
the surfaces. 

Some tests were carried out in 1904 at the Saint Louis Exposi- 
tion on locomotives, using grease and oil. A consolidated-type 
locomotive, 22 by 28 in., eight-wheel coupled, two wheels in front 
(2-8), developing a maximum power of 1,000 to 1,100 hp., 
showed a frictional loss as follows: 

Oil at 15 miles per hour: 61 hp. 

at 30 miles per hour : 107 hp. 
Orease at 26.6 miles per hour: 224 hp. 


A Pennsylvania consolidated-type locomotive, developing a 
maximum power of 1,000 to 1,100 hp., consumed in friction alone 
when using grease throughout: 

With grease at 15 miles per hour: 132 hp. 
at 30 miles per hour: 224 hp. 

It was demonstrated that wear of axles and crankpins was 
greater with grease than with oil and that there was not much 
difference in the cost of lubrication, the consumption with grease 
being approximately 450 miles per pound of lubricant. 

Outside the United States grease has not been favored for 
locomotive lubrication; in Europe oil is used everywhere in pref- 
erence to it. 

In Great Britain grease is, however, still used for lubricating 
colliery trucks and freight wagons, but this practice is rapidly 
dying out in favor of oil. 

The grease is placed in a cavity formed in the top of the axle 
box. Large openings in the bottom of this cavity communicate 
with similar openings in the brass, and under the influence of 
frictional heat the grease gradually melts and lubricates. The 
friction is high; the boxes are often neglected, lids are torn off, 
and the grease cavities contaminated with dirt, water, etc. 
Altogether the results are such that the sooner this form of 
lubrication is done away with in favor of oil lubrication the 

On page 27 will be found some information about the manu- 
facture and constituents of such railway-wagon greases. 

Railway Oils. The character of railway oils is governed to a 
large extent by the climatic conditions. In the tropics the oil 
is exposed to very high temperatures during the day and quite 
low temperatures during the night. Long-distance trains going 
from a warm low-lying country into a cold mountainous district 
will find themselves exposed to widely varying temperature con- 
ditions during their journey. In temperate climates the same 
conditions exist except that the differences between the day and 
night temperatures are smaller; still, the variation in temperature 
may be quite considerable. For example, the Scottish express 
trains running between London and Scotland will meet tempera- 
tures in the North very appreciably lower than those in the 


These conditions call for oils with low setting points in order 
that they may feed as uniformly as possible and with certainty 
through the oil siphons and other feeding appliances. 

On the other hand, once the oil has entered the bearing sur- 
faces it is exposed to considerable pressure and high temperature, 
so that it must possess great oiliness at the bearing temperature. 
In brief, railway oils must have viscosities that are not unduly 
influenced by great variations in temperature. The oils that 
best satisfy these requirements are mixtures of nonparaffin-base 
mineral oils with setting points in the neighborhood of 0F. 
mixed with from 10 to 25 per cent or even more of a suitable 
fixed oil. Mineral oil of the character described will give fluidity 
in the cold, and the admixture of fixed oil has the effect of main- 
taining great oiliness and viscosity at high temperatures. 

The admixture of fixed oil serves another purpose in the case 
of locomotive-engine oil, in that it prevents the oil from being 
washed away from the bearing surfaces by the steam which 
escapes from the piston rod and valve-rod gland, the condensed 
steam producing a " lather " on the guides and other parts. 

The setting points required for the blended oil can be deter- 
mined only on the road, although siphoning tests may be carried 
out in the laboratory indicating the siphoning and the capillary 
power of the oil at different temperatures, including the lowest 
temperatures to which the oil will be exposed during service. 
Such siphoning tests are not much used by railways, and yet 
they are of the greatest importance. 

Oils differ very considerably in their ability to siphon, and, 
furthermore, the quality of wool on the market varies very 
considerably in its siphoning qualities. In the case of siphon 
oilers, the wool that will give the greatest siphoning effect for 
the class of oil in use is the most desirable. In the case of pad 
oilers, which are fixed below the axles and lift the oil from the 
bottom of the box, the ability of the pad and its feeders to draw 
the oil and hold it is most important. It will be found that the 
quality of wool required for the two purposes is different. Wool 
or cotton that will lift the oil a considerable distance and hold 
it there will not easily deliver it to the journal, nor will it have 
good siphoning qualities when used in a siphon oil cup. 

As regards the viscosity of railway oils, it is always desirable 
that it should alter as little as possible per degree Fahrenheit. 


As a rule, the more fluid the oil the quicker will it feed through 
the lubricating appliances; and consequently if the oil varies 
greatly in viscosity with a varying temperature, the feed will be 
irregular and wasteful. When comparing oils for change in 
viscosity due to increase in temperature, the oils least affected 
at the bearing temperatures are the free-flowing vegetable or 
animal oils, while mineral lubricating oils made from either 
paraffin-base crudes or asphaltic crudes are distinctly inferior 
in this respect. When the running temperatures are low, 
approaching freezing point, the comparison may fall out differ- 
ently, as most vegetable and animal oils (as well as paraffin-base 
lubricating oils) have a poor cold test, whereas asphaltic-base oils 
still flow freely. 

The selection of the right quality of vegetable or animal oil is 
very important, because unsuitable fixed oils usually become acid 
during use and have a strong tendency to oxidize and produce 
gummy deposits. The acidity has an effect on the bearing 
metal, and that, in connection with the gumminess produced 
by the oil, attracts and fixes the dust and dirt that enter the 
bearing. As a result, the oiling pads or oiling waste or the oil 
siphons become choked and more or less inoperative, because of 
the deposit. 

The fixed oils used for compounding locomotive-engine oils 
may be rape oil, olive oil, or whale oil or mixtures of these; rape 
oil and whale oil are usually used in the form of blown oils, blown 
to a viscosity of 400 to 720 sec. Saybolt at 212F., and the per- 
centage ranges from 10 to 25, the same as for marine-engine oils; 
in fact, the character of the oils is very similar. 

Car oils are usually dark lubricating oils, containing less 
than 3 per cent of asphaltic matter and preferably compounded, 
although not to the same extent as locomotive-engine oils, as the 
bearing pressures that they have to withstand are much less. 

Car oils are preferably compounded with 8 to 12 per cent 
of animal oil (blown vegetable oils are apt to clog the pads). 
They are often used straight, i.e., not compounded, on account 
of the lower price per gallon. 

The specifications on page 291 are typical of locomotive-engine 
oils and car oils. 

In exceptionally cold climates lower setting points may be 
required; and when locomotive bearings are abnormally loaded, 






at 50C. 

Per cent 



Locomotive engine, 

summer grade . . . 




15 to 25 


Dark red 

winter grade .... 




"10 to 20 

Dark red 

Car, summer grade 




8 to 12 



winter grade .... 




5 to 10 


* See table, p. 57. 

a greater percentage of compound than recommended above 
may be needed, even to the extent of using pure rape or pure 
castor oil. Pure castor has here the advantage over other fixed 
oils of possessing an excellent cold test, which under great varia- 
tions in temperature is of great value. 



Streetcars are nearly always driven by electric motors but 
are occasionally operated by cables traveling below the streets, 
e.g.j the cable trams in Edinburgh. 

The important parts requiring lubrication are the axles, the 
motor, and the gearing. 

Axle Boxes. The construction and lubrication are often very 
similar to railway practice. One meets all sorts of combinations 
of siphon oiling (from the top), pad oilers, or oily waste packing 
(from below), the development being distinctly in favor of the 
last-mentioned oiling methods. 

The Armstrong and other pad oilers are widely used, but 
unfortunately many oil wells are made too small, so that it is 
difficult to fit the pads, and the wells contain too little oil. 

A very unsatisfactory combination of grease and oil lubrica- 
tion is not infrequently used. The oil is fed from below, and the 
grease, filling a cavity in the brass, acts as reserve lubricant. 
The trouble is that the grease becomes softened by the oil film 
on the journal and in time gets worked into the pad oiler below 
the axle, choking the pad and making it inoperative. 

With grease alone, the friction and wear are much greater 
than with oil, and the necessary period of oiling and inspection 
of the cars varies from once a day to twice a week, whereas 
with oil an inspection once every 2 to 6 weeks represents current 

The axle boxes are usually fitted with dust guards. This is 
important, to keep out not only the dust but also water, as, on 
rainy days and when the tracks are not properly drained, the 
wheels throw the water about; and if it gets inside the bearings 
in any quantity, trouble is sure to follow. 

Motor Bearings. Ring oiling is not uncommonly employed, 
and when suitable shapes of rings are employed (see " Ring-oiling 
Bearings/' page 162) the rings will run at such a speed that no oil 




spray is formed, and yet sufficient but not too much oil will be 
conveyed to the journal. 

Much trouble has, however, been experienced with ring oiling 
on electric cars, the oil escaping from the bearings and getting on 
to the commutator and rotor. 

Pad oilers are gaining in favor both for motor bearings and for 
suspension bearings, as they are very reliable in feeding the oil 
and do not overlubricate the journal. The pad must be placed 
so that it rests on the journal in a position where the oil can easily 
wedge its way in between the bearing surfaces (Fig. 102). From 

FIG. 102. Pad oiler. 

FIG. 103. Waste oiling. 

the pad a number of woollen siphon strands reach down into the 
oil well, which may hold a large amount of oil, or, if it is small, 
the oil should be fed continuously to the well from an oil cup 
placed in a suitable position. 

Oil-soaked waste is also used to some extent, feeding through an 
opening in the side of the brass, as shown in Fig. 103. The open- 
ing may be rectangular, with all sides well chamfered on the 
inside where the oil is to enter the bearing, and from each corner a 
shallow oil groove has been found advantageous to distribute 
the oil, on account of the rather sparing oil supply. 

Oil is added at intervals to the oil-soaked waste in the way 
indicated; in one case 1 pt. of oil had to be added every 120 to 
160 miles for a 5>- by 10-in. journal running 1,100 to 1 ; 600 r.p.m., 
the weight of the rotor being 2 tons, 



FIG. 104. Diaphragm-circulation oiling. 


FIG. 105. Reversible rotary pump. 


An interesting method of circulation oiling has been used for 
the motor bearings on a south-of-England electric railway, as 
shown in Fig. 104. The oil pump (1) pumps the oil in the same 
direction independent of the direction of its rotation, as will be 
seen from the detail drawing (Fig. 105). The oil is forced to the 
diaphragm plate (2) which has one, two, or three 1-mm. holes, 
through which a small amount of oil is constantly delivered to 
the bearing, the greater portion continuing its way to the suc- 
tion-joint box (3), where it joins the return oil from the bearing 
and finally reenters the oil pump. Each motor bearing has its 
own independent pump-supply, delivery, and return pipes. 

The wear of motor armature bearings on British streetcars 
ranges from 5,000 to 50,000 miles per ^le i* 1 - vertical wear, the 
wear of the suspension bearings being rather less; the average 
life of motor bearings appears to be 10,000 to 12,000 miles. 

The reason for such large wear as compared with stationary 
motors is the effect of fine hard grit and dust (wear from pave- 
ment, etc.) which are whirled up by the wheels and enter the 

During late years, considerable progress has been made in 
employing ball and roller bearings both for motors and for axle 

Gear Wheels. Most gear wheels are enclosed in a casing and 
use some kind of thin gear grease. The results are always inferior 
to those obtained with gear oil, but of course the gear case, if oil 
is to be used, must be as oiltight as possible. 

With grease or grease and oil, the life of the gear wheels may be 
from 50,000 to 200,000, whereas with oil the gears last consider- 
ably longer. 

The pinion wheels do not last so long as the gear wheels, but 
also here the use of oil is conducive to longer life. 

Oils. The oils used for lubricating the axle boxes of electric 
streetcars and railway cars are usually lower in viscosity than 
those used in railway practice, because the bearing pressures and 
conditions generally are not nearly so severe. Bearing oils 3 and 
4 (see page 135) represent oils that may be recommended for elec- 
tric streetcars, and bearing oils 4 and 5 are recommanded for 
electric railway cars. All of these oils should preferably be 
compounded with not more than 10 per cent of a nongumming 


animal oil, and in cold climates a low setting point would be 

The oils for motor and suspension bearings should be of a 
rather higher viscosity, as they are exposed to high temperatures 
(commutator heat) or pressure (from pinion wheel). 

Bearings oils 5 or 6 may be recommended and may with advan- 
tage be compounded when the conditions are severe. 

As to gear lubricants, the same oils as are used for the motor 
and suspension bearings can be used when the gear case is rea- 
sonably oiltight. When a more viscous lubricant is required, 
mixtures of oil and gear grease in suitable proportions, so that the 
mixture is not unnecessarily heavy, will form the best solution. 

Wheel-flange Lubrication. For electric locomotives which 
have to negotiate many curves, e.g., the electric locomotive serv- 
ice through the Saint Clair Tunnel, Switzerland, wheel-flange 
lubricators have given excellent service. The oil is contained in 
an airtight receptacle of 1-qt. capacity, whence it is led to the 
wheel flanges by pipes and sprayed upon the flanges by jets of 
air. The air is supplied through a J^-in. pipe, which is connected 
to the oil receptacle above the surface of the oil. A branch of 
this pipe is connected to the oil-delivery pipe which leads to the 
flanges. The air is controlled by an electric push button, so that 
the lubricant is applied only when needed, as on curves. This 
apparatus has been in successful operation since July 10, 1910. 
The six electric locomotives to which it has been applied haul 
1,000-ton trains up and down 2 per cent gradients on which flange 
wear has been rather heavy, owing to the many curves and the 
rather low center of gravity of the locomotives. Lubrication of 
the flanges has so improved conditions that 50,000 miles and more 
are now run between wheel tire turnings. This means that the 
wheels can be removed for turning at the same time that the 
armature is removed for commutator dressing. The former 
mileage made between tire turnings was from 12,000 to 25,000 
miles. Filtered reclaimed armature-bearing oil is the lubricant 



The long main lines of shafting used for power transmission 
are called "line shafting. 11 . Counter shafting is driven from the 
line shafting and operates the various machines by fast and loose 
pulleys or by clutches. 

The speed of shafting ranges from 120 to 450 r.p.m.; the 
dia;,meter of line shafting usually 
ranges from 2% to 6 in. ; of counter- 
shafting, from 1 to 2^ in. 

Many bearings on countershaft- 
ing and small-diameter line shaft- 
ing are hand oiled or oiled by 
glass-bottle oilers. Line-shafting 
bearings are seldom hand oiled; 
they are usually bottle oiled, and 
modern shafting is frequently ring 
oiled. Ball and roller bearings are 
also coming into prominence for 
quick-speed line shafting. 

Heavy large-diameter shafting 
bearings, e.g., many second-motion 
shaft (jackshaft) bearings, develop 
so much heat that they can be kept 
cool only by a circulation-oiling 

System. FlQ 106 Screw-circulation oiling. 

Figure 106 shows a simple form. 

The screw can be lifted right out for examination by taking hold 
of the knob. 

Figure 107 shows a more elaborate system with three oil feeds 
from the oil box. The drawing will need no explanation. 

The power required to drive the line and countershrifting in a 
mill or shop is always a considerable percentage of the total load. 
In textile mills it ranges from 20 to 60 per cent; in engineering 
workshops, from 20 to 75 per cent. Whether more or less 




machines are in operation, the shafting load is always of the same 
magnitude, and it is not too much to say that in most existing 
factories or works an average of 10 per cent could be saved in the 
shafting load by introducing better lubricants, and another 
10 per cent by regular attention to keeping the shafting in perfect 
alignment. Losses from poor alignment and from unsuitable oils 
frequently occur simultaneously. Poor alignment often means 

FIG. 107. Pump-circulation oiling. 

that certain bearings heat because of the extra load; instead of 
the bearings' being adjusted, the oil gets the blame, and a more 
viscous shafting oil is introduced, which " cools " the bearings 
inclined to heat and at the same time adds 10 to 25 per cent of 
extra fluid friction to all the other bearings. If bearings are kept 
in good alignment, low-viscosity shafting oils can be used, and a 
considerable saving in power obtained (see remarks, page 326, 
regarding shafting in textile mills). 

Where electric driving is employed, it is a simple matter to 
take the shafting load every 3 or 6 months, as a check on the 



efficiency. With steam plants, the indicated horsepower may be 
recorded, or the number of revolutions of the flywheel and the 
time taken before it comes to rest from full normal speed, after 
steam has been shut off. 

Shafting bearings should be provided with save-alls to prevent 
dripping of lubricant. Oil creeping along the shaft, when it does 
occur, is usually only toward one 
side of the bearing and may be 
overcome, as shown in Fig. 108, 
by an oil thrower (1) and splash 
guard (2). The oil drops from 
the splash guard into the save-all 
(3). (As regards ring-oiling bear- 
ings, see page 161.) 

Ball and roller bearings save a 
great deal of power; a type of 
roller bearing very suitable for 
line shafting is the Hyatt flexible 
roller bearing (Fig. 49, page 181) 
which gives a coefficient of fric- 
tion of 0.005 to 0.008, whereas 
ball-shafting bearings give a 
coefficient of friction of 0.002 to 
0.003. Good alignment is essen- 
tial with ball and roller bearings, 
more so than with plain bearings, 
an exception being the Skefko ball bearing. The following fig- 
ures indicate the coefficient of friction that may be expected 
for different methods of lubrication in connection with shafting 
bearings : 

Coefficient of Friction 

Ball bearings 0.002 to 0.003 

Roller bearings 0.005 to 0.008 

Ring-oiling bearings 0.010 to 0.015 

Bottle oiling, siphon oiling . 02 to . 04 

Hand oiling 0.04 to 0. 15 

The great savings in power that follow the introduction of 
high-class shafting bearings is better realized on the Continent of 
Europe than elsewhere; in Great Britain and the United States 
conditions of shafting are much behind Continental practice. 

FIG. 108. Shafting oil thrower. 



Lubrication. Most shafting bearings are lubricated by oil; 
as mentioned elsewhere, shafting in weaving sheds is frequently 
lubricated by grease applied through gravity grease cups/ spring 
grease cups, or applied direct to the shaft. Stauffer cups are not 
used, because they must be given a turn every day or two, 
while the other methods are more or less automatic in action and 
require attention only at long intervals. 

The waste in power by applying grease, as compared with oil, 
ranges from 5 to 20 per cent of the shafting load, according to the 
fluidity and quality of the grease and the speed of the shafting. 

The better the lubricating system the lower viscosity oil can be 
used, and the lower the friction. 

For hand oiling, oils compounded with, say, 5 per cent of a 
nongumming fatty oil will last longer and give better results than 


Shafting oil 

at 50C. 

Bearing oil 2* 
Bearing oil 3 . ... 


For most moderate- and high-speed shaft- 
ing and countershafting in good align- 
ment and condition and with reasonably 
good lubricating appliances 
This oil is usually too thin for hand-oiled 
For slow- or moderate-speed light and 

Bearing oil 4 


medium shafting and countershafting in 
good or moderate condition and with good 
or moderate lubricating appliances 
Also for hand-oiled bearings on counter- 
For slow- or moderate-SDeed heavy shaf tine 

NOTE. For lubricants for ball and roller bearings, see page 193. 

Shafting greases. . . 

Grease should be of as light a consist- 
ency and as low a melting point as practi- 
cable, without incurring undue waste of 

The mineral oil used in the grease should be 
of viscosity similar to that of the oil that 
would prove suitable if the bearings were 
arranged to use oil instead of grease 

* For bearing oils, see p. 135. 


straight mineral oils. For bottle oilers, straight mineral oils 
should be used to ensure the needles' keeping clean and in work- 
ing order. Oils for ring-oiling bearings and ball bearings should 
also be straight mineral. 

The chart on page 300 is a rough guide for selecting the correct 
grade of shafting oil. 



Machine tools are machines such as lathes, shapers, and boring, 
drilling, milling, planing, and grinding machines, the speeds 
ranging from quite low on large lathes and planers to very high 
up to 10,000 to 30,000 r.p.m. for modern grinders. 

A great many bearings on most machine tools are hand oiled, 
the speeds or pressures being low. The oil holes should prefer- 
ably be protected by a cover. Figure 109 shows a typical oil- 
hole cover; the lid (1) is turned, disclosing the oiling hole (2); 
the lid, by means of an internal spring, may be made to turn back 

FIG. 109. Oil-hole cover. FIG. 110. Ball valve FIG. 111. Oil-hole 

and felt chamber. protector. 

Hand-oiling arrangements. 

automatically and cover the hole after the oiling operation. 
Figure 110 shows a hand-oiling arrangement with ball valve (1) 
and felt chamber (2). Felt, wool, or worsted yarn may be used 
in the chamber and serves to feed the oil more uniformly to the 
bearing in between oilings. With a rise in temperature more oil 
is liberated, so that such an arrangement is a great improvement 
over the ordinary oil hole without felt. 

Figure 111 shows a simple oil-hole protector, consisting of a 
cup, the shank of which is split in three parts which grip the oil 
hole as the cup is pressed into position. The cup and shank are 
filled with felt, which acts in the same way as the felt in Fig. 110. 

In many modern machine tools, felt-pad arrangements are 
made use of to a considerable extent. Figure 112 shows an 




arrangement used by Brown and Sharpe for the bearings of 
internal-grinding spindles. The oil soaks through the felt and 


FIG. 112. Felt-oiling arrangement for grinder spindle. 

enters the bearing through the passage shown. 

In many bearings large recesses are cored out around 
the spindle boxes in the middle 
and fitted With felt pads, which 
are pressed gently against the 
revolving spindle by means of 
light feather or spiral springs. 

Figure 113 shows two types 
of pads; when in use they are 
both placed below the spindles 
in a well partly filled with oil, 
which is replenished from time 
to time through an oil-filling 
hole at the top communicating 
with the oil well. Right- and 
left-hand spiral grooves, as 
shown in Fig. 114, are excellent 
for distributing the oil toward the bearing ends, where fine V 
threads on the spindle cut in the opposite direction t&nd to pre- 
vent leakage and have proved very efficient iii this respect. 

Bearings that require a fair amount of oil may be supplied by 
small siphon oil cups or drop-feed oilers; occasionally, ring-oiling 


FIG. 113. Spring felt pads. 



bearings are employed. In some recent designs a circulation- 
oiling system is employed, a pump delivering the oil to a distri- 
buting box, whence oil is guided to the various bearings and 
gears and finally returns to the pump reservoir. Grease is sel- 
dom used for machine tools, except in ball bearings, which are 
now widely used, especially as vertical thrust bearings for drill 
spindles, heavy revolving tables, etc. 


Fio. 114. Spiral oil grooves for grinder spindle. 

The lubrication of lathe saddles, ram slides of shaping machines, 
and flat or V-shaped slides of planing machines is receiving more 
attention nowadays. Instead of the surfaces' merely being 
flooded by an oilcan, most of the oil being wasted to no good pur- 
pose, some modern machines have felt-pad insertions in the sliding 
member. The felt pads are kept soaked with oil, being hand 
oiled through oil passages from above, and keep the large sur- 
faces economically and fairly well lubricated. In some V-grooved 
slides, V-shaped wheels are placed in the stationary slides; the 
wheels are partly immersed in oil and are forced gently against 
the moving slide which they lubricate. The felt-pad arrange- 
ment is probably equal to if not more efficient than the revolving 



Apart from high-speed machine tools, the majority of bearings 
in machine tools are only poorly lubricated at the best of times, 
and the coefficient of friction is high. Slightly compounded oils 
are therefore preferable to straight mineral oils, as they have 
greater oiliness. The low-viscosity oils, which are (or ought to 
be) used for high-speed tools like grinders, need not be com- 
pounded, as the friction depends upon the viscosity of the oil and 
not on its oiliness. 

Exposed in thin films to the oxidizing influence of air and fine 
metallic dust, the oil which invariably creeps all over the machine 
tools in time oxidizes and stains or tarnishes the bright surfaces, 
particularly in machine shops exposed to bright light or sunlight. 

In all mineral oil there are certain complex unsaturated hydro- 
carbons, coloring matter, etc., which are easily oxidized and which 
are the cause of the brown, thin, tenacious films just referred to. 

Pale mineral oils are less apt to cause tarnishing than dark- 
colored oils, and it is a great help to have a small percentage of 
animal oil, say 6 per cent of lard oil, mixed with the mineral oil. 
The admixture of animal oil has a marked effect in preventing the 
oxidized matter from forming a film and makes it quite easy to 
wipe the surfaces clean. 

An admixture of a vegetable oil will have the opposite effect ; it 
helps to cement the oxidized matter together and makes it more 
difficult to keep the bright surfaces on the machines clean. 

Oil of Three Viscosities Are Required as Follows: 


at 50C. 

Type of machine tools 

Bearing oil 1 * 


For very high-speed machines, as grinders 

(Straight mineral) 
Bearing oil 2 


For all moderate- or high-speed machine 

(Preferably pale 
and compounded 
with 6 per cent of 
lard oil) 
Bearing oil 4 


tools of every description, except 

For all slow- or moderate-speed, heavy 

(Preferably pale 
and compounded 
with 6 per cent of 
lard oil) 

machine tools, for gear chanpbers, etc. 

* For bearing oils, see p. 135. 



The textile industries, comprising the cotton, woolen, worsted, 
silk, rayon, flax, hemp, and jute industries, are all highly specialized 
and employ such a variety of machinery that it is impossible 
inside a few pages to give even an outline of the principal types 
and their uses. 

Characteristic of most of the machines is that the amount of 
power actually used in doing useful work, i.e., in handling the 
fibers or material itself, is small and that by far the greater por- 
tion of power is consumed by the friction of numerous spindles 
or shafts often revolving at high speeds and usually only lightly 

Great improvements have taken place so far as the mechanical 
construction and lubricating arrangements are concerned, and 
the author will endeavor in the following pages to point out 
some of the important features. While considerable attention 
has been paid to the selection of suitable oils, yet very great 
power reductions can be accomplished in practically all existing 
mills by the introduction of such oils as will be mentioned later 

The subject will be divided into four sections, viz.: 

1. Preparing. 

2. Spinning. 

3. Weaving. 

4. Bleaching, Dyeing, Printing, Finishing. 


Openers and Scutchers (Used for Cotton Only). Openers 
and scutchers are very similar in action; they open and loosen 
the fibers of cotton by quickly revolving beaters; the cotton fluff 
thus formed is blown a certain distance and again gathered 
together, forming a soft thick sheet of cotton called a "lap." In 
this process the cotton fibers are cleaned from dirt and grit, pass- 
ing first through the openers and next through the scutchers. 



There are quickly revolving spindles in these machines, the lubri- 
cation of which is important. By feeling these bearings, an 
expert can always get an idea of the quality of the spindle oil 
used in a mill; if they run excessively warm, the oil in use is 
probably too viscous, assuming, of course, that the bearings are 
in good condition mechanically. 

The high-speed bearings are either oiled by bottle oilers or, 
preferably, ring oiled. 

The room in which the openers and scutchers are placed is 
called the " blowing room," or " scutching room." 

Washing and Drying Machines (Used Only for Wool and 
Worsted). Wool-washing and -drying machines do not present 
any lubrication features of interest, except that in some mills 
hydroextractors are used for " whizzing" the wool before it passes 
into the drying machines for the final drying. 

These hydroextractors are of the same type as those used for 
recovering oil from waste (Fig. 231, page 612), and unless they 
have ball bearings or Michell bearings they require oils of great 
oiliness much more viscous than the spindle oils used in the mill. 
Hydroextractors are usually driven direct by a small steam 
engine or steam turbine. 

Preparer Gill Boxes (Used for Wool, Worsted, Flax, Hemp, 
Jute, and Waste Silk). These machines comb open the fibers, 
lay them parallel, and deliver them in the form of a continuous 
"end," or lap. The material always has to pass through several 
sets of gill boxes. 

The last preparer gill box in the series is called the can gill box 
and is shown in Fig. 115. The lap (1) enters the back rollers 
(2) and is drawn between the front rollers (3) and delivered 
through the slowly revolving funnel (4) as a continuous sliver into 
the can (5). Between the front and back rollers the fibers are 
combed by the fast-moving fallers (6) which rest with their ends 
on slides and are pushed to the right by means of square-threaded 
screws; they fall at the end and are returned quickly by bottom 
screws (revolving in the opposite direction) to be raised again 
into position just behind the front rollers. 

The fallers, slides, screws, etc., wear rather quickly /and good 
lubrication is therefore extremely important, particularly when 
working with dusty fibers, such as jute and hemp. The dust, 
which is composed of earthy particles, also small pieces of 


woody and fibrous matter, contaminates the oil on all rubbing 

If when leaving the gill boxes the fibers (such as wool) go to 
the carding machines, they must be oiled. The oiling should not 
be done in the first, second, or third gill boxes but preferably 
in the can gill box. One method of oiling is shown in Fig. 115. 

FIG. 116. Preparer gill box. 

A circular brush (7) revolves in the oil trough (8). When the 
bristles of the brush pass the blade (9) they shower or spray 
the hot oil on to the fibers of the wool as they pass through 
the machine. 

Carding Machines (Used for All Short Fibers, Not for Long 
Worsted and Long Silk). The carding operations remove all 
impurities and arrange the fibers parallel, delivering the material 
in the form of sliver. 

The soft laps coming from the blowing room enter the card- 
ing machine and are broken up by the revolving cards, being 
delivered from a large carding drum to smaller carding drums, 
which return the fibers to the main drum; finally, the fibers are 
removed in the form of a thin veil from the last drum by means of 
a quickly oscillating stripping comb. The veil is gatheped 
together through a trumpet, passes a pair of rollers, and is 
delivered as sliver into a card can. 

The bearings for the stripping comb are placed in so-called 
"stripping-comb boxes/' which contain a bath of oil and in which 
cams operate and give motion to the stripping comb. These 
stripping-comb boxes are always rather warm and indicate the 


quality of the spindle oil. The numerous bearings on the carding 
machines require to be well oiled. Several of the spindles sup- 
porting the smaller carding drums have an endwise oscillating 
motion, tending to scrape off the oil film. 

There are usually two or more sets of cards before the sliver 
is passed on to the drawing department. 

Short wool does not leave the cards as sliver, but, before 
going to the drawing frames, it is passed from the cards straight 
into so-called " condensers "; the wool enters these as a thin soft 
sheet and is divided into a number of strips, which are rolled into 
coarse threads, suitable for coarse spinning, which is the next 

Combers (Used for All Long Fibers, Only Rarely for Cotton). 
Long wool, worsted, flax, and other long fibers are not carded 
but pass through combers. There are many types of combers, 
but the object in them all is the same, i.e., to straighten the fibers 
and separate the short from the long ones. 

Most parts of these machines, such as revolving tables and 
drawing-off rollers, revolve slowly and require a rather viscous oil, 
but the " dabbing brushes " have a quick motion and should 
preferably use thin spindle oil. Modern dabbing motions are 
enclosed in a chamber containing oil to ensure continuous lubri- 
cation, and a speed of 800 to 1,200 dabs per minute can be 
obtained without unreasonable vibration. 

The slowly revolving tables "circles" are often supported 
by balls placed in ball races. These races become very hot 
when the circles are steam heated, and the oil will carbonize and 
gum unless the oil manufacturer has kept this condition in 
mind and selected a "noncarbonizing" oil. Some circles are 
supported by large rollers, which revolve and dip into oil reser- 
voirs and are thus kept continuously oiled. 

Drawing Frames. The drawing frame receives thick "slivers" 
of fibers and attenuates them by the so-called "drawing" 

The frame consists essentially of several sets of rollers, each 
successive pair revolving at a greater speed than the previous 
pair. The top rollers are weighted, and the bottom rollers fluted 
to grip the fibers tightly. 

When drawing material like wool or worsted the rollers are 
heavily pressed together, and a specially viscous oil is required; 



with cotton the rollers are not so heavily loaded, and they are 
easier to lubricate. Care must be taken not to overlubricate, as 
if the oil gets on the rollers it will produce oil stains on the yarn. 
The bearing keeps for the roller bearings should preferably be 
fitted with flannel layers inside, which have the effect of holding 
and distributing the oil all over the bearing surfaces and keeping 
the dust out. 

Slabbing, Intermediate, and Roving Frames. Slubbing, inter- 
mediate, and roving frames are used for producing coarse thread 
from the sliver coming from the drawing frames, the sliver 

FIG. 116. Slubbing frame. 

passing through several of these frames in the order indicated. 
Slubbing and intermediate frames are used only in cotton mills; 
for other fibers only roving frames are used. 

All of these frames are flyer frames and very similar in 

Figure 116 shows a slubbing frame. The sliver passes from the 
can (1) through draft rollers (2), through the hollow arm of the 
flyer (3), and is wound on to the bobbin (4) driven by skew wheels 



(5) at a slightly lower speed than the flyer, which is driven by 
skew wheels (6). The bobbin together with its wheel drive is 
continuously lifted and lowered during the operation. 

The spindle has a footstep bearing and a neck bearing, both 
usually oiled by hand. 


The object of spinning is to draw out and twist the coarse 
thread received from the preparing department and produce a 
more or less finely spun yarn. There are four main types of 
frames, viz., ring, flyer, cap, and mule frames. 

Ring Frames. Figures 117 and 118 show this type of frame and 
spindle. The thread is drawn by the draft rollers (2) from the 

FIG. 117. Ring frame. 

FIG. 118. Flexible ring 

bobbins (1) and delivered through the eye (3) to the bobbin (4). 
The bobbin (4) is driven from the tin roller (5), pulls the thread 
through the "traveler 77 (6), and continuously winds up the 
yarn. The traveler revolves on the ring (7) fixed on the lifter (8). 



The bobbin is fixed on the spindle (9) which is surrounded by 
a sleeve and immersed in an oil bath. Several holes are provided 
in the sleeve which allow the oil to enter freely at the bottom 
and the side. Some of the oil rises along the spindle, overflows 
at the top, and returns through a vertical passage to the oil 
reservoir at the bottom. 

The casing and oil reservoir in which the spindle revolves is 
called the bolster. It will be noticed that the spindle sleeve is 
provided with a spring which will allow it slight lateral move- 
ments in relation to the bolster. 

Make-up oil is added at intervals through the oil well (10) 
which communicates with the bottom oil reservoir and is pro- 
tected from dust owing to the shape of the driving whorl (11). 

The so-called "Rabbeth spindles " are now going out of use; 
they are similar to Fig. 118 except that the spindle sleeve is 

FIG. 119. Ring-spindle oihran. 

FIG. 120. Ring-spindle oiling 

rigidly fixed in the bolster. They cannot be operated at speeds 
higher than 6,000 r.p.m., as they are then inclined to throw off the 
bobbins. The flexible-type ring spindles are operated smoothly 
at speeds ranging from 6,000 to 11,000 r.p.m. notwithstanding 
slight unevenness in the driving bands. 

When a new frame is being started the oil should be pumped out 
after 2 days' working, and fresh oil introduced. The oil should 
be renewed after a week's run and again after 4 weeks' further 
running. Current practice for oiling frames afterward is to add 
a little fresh oil every 3 months to the oil wells and to empty 
them for cleaning and recharging once every 12 months. 

Figure 119 shows an oilcan for refilling spindle baths. The 
measure (1) is lowered in the tube shown, filled with oil, and, when 


lifted, tips over its contents into the spout (2), which pours the 
oil into the spindle bolster. 

Another type of oilcan is also used for this purpose, in which 
there is a plunger pump which is pressed down by the thumb. 
An adjusting screw is fixed below the thumbpiece by means of 
which the amount of each discharge can be adjusted. The 
delivery spout may have a sight-feed arrangement to indicate 
that the pump is in working order. 

By connecting all the bolsters to a horizontal oil pipe (Fig. 
120) and having an oil-filling vessel (1) at the end, the oil level 
is correctly maintained for all spindles. It cannot become too 
high, because of the overflow (2) which discharges excess oil into 
the small oil receiver (3). The system can be drained by remov- 
ing the drain plug (4). 

While this system is excellent for preventing shortage of oil 
in the bearings, it carries with it the danger of forgetting to over- 
haul and clean the spindles, which is important and ought to be 
done at least once per annum. 

Flyer Frames (Used for All Fibers). Figure 121 illustrates a 
typical flyer spindle. The flyer (1) revolves and lays the yarn 
on the bobbin (2), which is lifted and lowered by the lifter (3). 
The spindle is supported by a neck bearing (4) in the rail (5) and 
a footstep bearing (6). 

The small recess shown in the center is not often found in 
spindle-footstep bearings but is a great advantage; it prevents 
heating of the spindle tip and serves to collect dirt which other- 
wise would cause friction and wear. On very heavy spindles it 
would probably be beneficial to let the oil circulate, as indicated in 
Fig. 121, the action being the same as in ring-spindle footsteps. 

Flyers used for wet spinning (flax mills) should have their 
tops enclosed, as shown in Fig. 122, to prevent entrance of 
moisture, which causes rusting and makes it difficult to unscrew 
the flyers, unless a heavily compounded oil is used for oiling the 
spindle tops. 

Figure 122 shows a patent flyer spindle (the Bergmann spindle) 
used for spinning flax, hemp, and jute. The spindle is driven 
in the usual manner, but the whorl is in line with thf; footstep, 
so that the principal object of the neck bearing is to steady the 
spindle. The neck bearing is made very flexible by means of 
feather springs (1) and is covered with a lid to keep out dirt and 



fluff from the felt oil pad which keeps the spindle well oiled. The 
whorl protects the footstep from dirt, and in this type of footstep 
the oil may be arranged to circulate in the same way as in the 
footsteps of ring spindles. If the spindle is lifted by means of 
the whorl, the footstep bearing is disclosed for examination and 

The felt-pad arrangement here shown (2) and also used for 
many cap spindles (Fig. 123) ought to be much more widely 

FIG. 121. Flyer 

FIG. 122. Bergmann 

FIG. 123. Cap 
spindle with felt- 
pad oiling. 

used for neck bearings of flyer spindles; it is simple, efficient, 
and economical. 

An attempt has been made to introduce oil circulation for 
the neck bearings. The rail is hollowed out and forms an oil 
reservoir; the oil passes slowly through tiny openings in the neck 
collars into the neck bearings; by means of collars on the spindles 
below the rail, the oil is thrown off into dishes surrounding the 
spindle, returning through pipes to an oil reservoir, whence a 
pump takes the oil and delivers it into the rail. The oil thus 
circulates continuously. It should be drawn off every 3 months 
and filtered and can be used again, if of good quality. This 


arrangement is, however, rather complicated and not so fool- 
proof as the felt-pad arrangement. 

One type of flyer frame, the " Arnold Forster," has the spindles 
fitted with ball bearings and a self -lubricating felt pad to ensure 
smooth and easy running. 

Cap Frames (Used for Wool, Worsted, and Waste Silk). 
A typical cap spindle is illustrated in Fig. 123. The spindle 
(1) is stationary, and the cap (2) rests on its top. The bobbin is 
revolved by means of the whorl (3) operated by a driving band 
from the tin roller. The bobbin continuously winds up the yarn 
and pulls it over the bottom edge of the cap. The lifter (4) raises 
and lowers the bobbin, which slides with a long brass tube 
(5) on the spindle. 

Obviously, it is very important to oil this tube well; the felt- 
pad arrangement (6) is very efficient and economical, it being 
sufficient to oil the felt pad once every week or fortnight. In 
many cap spindles there is no felt pad, and the spindle is dabbed 
once or twice a day with an oily brush; this old-fashioned method 
means a higher oil consumption, more wear, and about 10 per 
cent higher power consumption. 

Mule Frames (Used for Cotton, Wool, and Waste Silk) 
(Fig. 124). The mule spindles (1) are placed on a movable car- 
riage (2) which during the spinning period moves to the left, 
while the draft rollers (3) draw the thread from the bobbins (4). 
When the carriage moves to the right, the yarn is wound on the 
spindles, the fallers (5) moving down into such positions as to 
guide the yarn correctly on to the spindles. 

Mule spindles have a neck bearing and a step bearing, the 
same as the flyer spindles, the only difference being that they 
are placed at an angle; the oil is therefore inclined to be thrown 
out of the footsteps. One method of minimize waste of oil due 
to this cause is to protect the footsteps, e.g., with Jagger's foot- 
step protector, shown in Fig. 125, which has proved very useful. 
It also protects the bearing from dirt and fluff, and during oiling 
it catches all oil; without protectors much oil often runs down 
the rail and is wasted. 

The neck bearings are oiled once, twice, or three times per 
day according to operating conditions and the class of oil in use. 
The footsteps are usually oiled the same number of times per 
week as the neck bearings are oiled per day. 



In the center of the mule is situated the headstock, from which 
all parts of the frame receive their motion, and it is regarded as 

FIG. 124. Mule frame. 

one of the most ingenious and complicated machines in the textile 

FIG. 125. Jagger's footstep protector. 

Driving bands are usually made of cotton and are affected 
by the moisture in the air. With most spinning frames the 



consumption of power varies approximately 1 per cent for every 
6 per cent variation in the relative humidity of the atmosphere 
in the spinning room. The higher the relative humidity the 
more the bands contract, and the higher the power consumption. 

With some modern frames, notably cap frames and jute 
spinning frames, the driving bands have their tension maintained 
uniform by means of weighted tension pulleys, as shown for a 
cap frame in Fig. 126. 

The tension need therefore never be any more than that 
required for driving the spindles at their correct speeds, and 

FIG. 126. Uniform belt tension arrangement. 

humidity has no influence on the power consumption. A higher 
spindle speed can obviously be obtained with this type of drive; 
and as the spindles are never subjected to excessive strains from 
the band pulls, their lubrication is easier; lower viscosity spindle 
oils can be employed with confidence; and the power consumption 
can then be considerably reduced as compared with frames 
employing the ordinary type of band drive. 

Thread, Twine, and Cord. In the treatment and irianufacture 
of thread, twine, and cord a variety of light machines are 
employed, such as doubling, winding, and gassing frames; reeling 
machines; twisting, twine, and cord machines; thread-polishing 


machines; and balling and spooling machines, the lubrication of 
which presents no striking features. 

Doubling frames have either flyer spindles or ring spindles. 
There are some self-acting doubling frames (twiners) very 
similar to mule frames. Some winding frames employ ring 

Rope-making machines are either vertical or horizontal, the 
former being used chiefly for large cables. 

From the lubrication point of view these machines, which 
often look ponderous and complicated, consist essentially of 
revolving bobbins and are not difficult to lubricate. 

Wool Oils and Batching Oils. Wool oils are used for lubri- 
cating the fibers preparatory to the carding operation. 

With all high-class wool the oil must at a later stage be com- 
pletely removed, as otherwise the yarn will not take the dye prop- 
erly. Olive oil is undoubtedly the best grade of wool oil. It is 
easily removed but is expensive, therefore only used for the 
highest class of material. Other fixed oils, such as nut oil and 
lard oil, are almost as good as olive oil but are also expensive. 
Wool oleins (produced from wool grease) and various fatty acids 
(oleic acids) are much used mixed either with a percentage of 
other fixed oil or with mineral oil, even up to 80 per cent of the 
latter. The lower the class of material and the more intense 
the scouring methods the more mineral oil can be used in the 
mixture, without running undue risk of having trouble in the 
dyeing of the yarn. The wool oil must never contain more than 
6 per cent of fatty acid, or 12 per cent of wool olein (which 
normally contains 50 per cent of free fatty acid), as more acid 
weakens the fibers and destroys the wires on the carding machines 
as well as the pins of the preparing and combing machines. 

Rape oil, cottonseed oil, and the like are not so suitable, 
as they oxidize and produce gum deposits in the machines. 

Some mineral oil 20 to 30 per cent should always be 
present in the wool oil wherever permissible, as its presence 
greatly reduces the well-known tendency that all fixed oils, 
particularly vegetable oils, have for spontaneous heating, which 
has been the cause of many outbreaks of fire. 

Batching oils are used for softening the fibers of flax, hemp, and 
jute. Low- viscosity mineral oils are generally used, and occa- 
sionally mixtures of whale oil and mineral oils. 



Winding, warping, and sizing machines prepare the yarn for 
the weaving process. The lubrication of these machines calls 
for no comment. 

Looms. There is an immense variety of looms, from small, 
quick-speed cotton or silk ones to large, slow-speed carpet looms. 

The function of all looms is to form a fabric by interlacing warp 
and weft threads ; there are three essential movements in a loom : 
shedding, picking, and beating up. 

Shedding is the operation of dividing the warp into two portions 
for insertion of the weft. 

Picking is the operation of passing the shuttle containing the 
weft through the opening formed in the warp. 

Beating up is performed by the reed and sley, which, through 
the action of cranks and connecting rods, advance and recede 
from the cloth after each "pick" in order to place the weft 
threads parallel with one another. 

Picking motions are called "overpicks" or "underpicks," 
according to whether the shuttle receives its motion from an arm 
placed above or below the sley. Overpick is generally used for 
fast-running looms, and most heavy slow-speed looms have the 
underpicking motion. This motion is cleaner, as oil is not 
required about its parts near the cloth, and is therefore preferable 
for white and light-colored goods, on which oil stains show up 
more than on dark-colored fabrics. 

The shuttle at the end of each journey is arrested by running 
into an "eye" made of buffalo hide and fixed in the shuttle box; 
the buffalo hide is steeped in neat's-foot oil to preserve it and to 
minimize wear of the shuttle nose. 

The shuttle gets its motion from a buffalo-hide "picker" slid- 
ing on the picker spindle and connected with the driving arm by 
means of a leather strap. The driving arm has a jerky motion 
which causes the picker to hit the shuttle hard and send it across 
the loom to the shuttle box on the other side. It may also be 
arranged in the form of a lever, which acts on the picker direct. 

The picker spindle is lubricated by dabbing it at intervals with 
an oily brush. A patent automatic picker-spindle lubricator is 
in use on overpick looms and consists of a small pad saturated 
with oil and carried by an arm which brings it into contact 


with the picker spindle at each forward movement of the sley 
and, on the return movement, again makes the pad recede, to 
give room for the passage of the picker. 

The danger of oil's getting on to the cloth increases with the 
speed of the loom. The speed is given in number of picks per 
minute and ranges from 240 for narrow looms and fine material 
down to 20 picks per minute for very coarse goods; for most 
woolen or worsted cloths the picks number from 60 to 70 per 

With quick-speed looms the cranks operating the reed and 
sley are apt to throw oil on to the fabric, particularly so when 
the bearings are overlubricated. 

In velvet looms the fabric is woven over a number of long 
"needles," which are continuously withdrawn from the finished 
portion and inserted again; in large velvet looms it is an advan- 
tage to oil these needles sparingly with " stainless " oil. 


Bleaching and dyeing departments employ comparatively little 
machinery requiring lubrication. The most important machines 
from our point of view are probably the hydroextractors. 

Printing machines (calico, thin woolen, linen, jute) are usually 
hand oiled, the same as other printing machines. 

The finishing processes are very varied. 

For cotton goods the main operations are singeing, raising, 
shearing, brushing, steaming, starching, calendering, impreg- 
nating, breaking down, damping, mangling, moir6ing, embossing, 
tentering and stretching, doubling, measuring and plaiting, 
marking, and pressing. 

For woolen and worsted cloth the main finishing operations are 
crabbing, scouring, milling, singeing, dyeing, raising, wet rolling, 
tentering, cutting, brushing, shrinking, pressing. 

Again here, hydroextractors are used after the dyeing process^ 
and most of the machines used up to this point are fairly heavy 
slow-speed machines, requiring a viscous oil for lubrication. In 
the scouring process any oil stains received during manufacture 
must be scoured out; in the subsequent operations extreme care 
must therefore be taken to avoid oil stains, and stainless oil 
should be used for lubrication in the last few stages, i.e., cutting, 


brushing, and shrinking. The pressing is generally done in a 
hydraulic press. 

For linen cloth the following finishing operations are used: crop- 
ping, washing, tentering, beetling, calendering, pressing. 

For jute cloth the finishing processes are as follows: damping, 
cropping, calendering, folding. 

The only machines calling for comment are the calenders, of 
which there are several forms, all consisting of several heavy 
rollers called " press bowls" placed horizontally in a strong frame 
and pressed against one another with more or less pressure 
either mechanically or hydraulically. 

The bearing brasses, top and bottom, should preferably touch 
the journals over an arc of only 90 to 120 deg., and the edges 
should be well chamfered to facilitate the entrance of the oil; 
when there are a number of bearings one above the other, the 
waste oil from one should be guided into the bearing just below, 
and so on. Some of the bowls are heated by steam or gas, and 
their journals become extremely hot so much so that oil cannot 
be used, and high-melting-point greases have to be employed. 
The wear of calender bearings is often very considerable. 


As most oiling in textile mills is hand oiling, it is extremely 
important to have the oilcans in good condition and see that they 
are maintained with small spout openings. Some oilers are 
inclined to cut off the ends of the spouts to make the oil flow more 
readily, and the result is a great waste of oil, as when a row of 
spindles is oiled the spaces between them are oiled as well as the 

FIG. 127. Oil-saving devices. 

spindles. The oilcans should be so adjusted that as the oiler 
goes along the frame at a regular speed, a drop of oil falls into 
each bearing. 


Figure 127 illustrates two methods of regulating the oil flow 
from the oilcan. The top illustration has an inside cone with a 
tiny opening, so that it is impossible to get a rapid feed of oil 
from the end of the oilcan spout. The cone cannot be interfered 
with by the operatives and can be made of any size according to 
the requirements. The bottom illustration shows the orifice of 
the spout itself, soldering a strong cap on to the end with an open- 
ing of, say, ^2 * n ' The drawback to this arrangement is that 
the operatives can easily cut off the cap, whereas they cannot 
interfere with the cone arrangement shown in the other 

It is a great advantage to have in each of the spinning rooms a 
small cabinet holding a few gallons of oil sufficient, say, for one 
week's consumption. The cabinets should be arranged with lids 
that can be padlocked. A small oilcan can be filled from the 
cabinet without waste, and the oil is always kept clean. Such 
small cabinets can be used for conveying the oil from the store- 
room into the various departments. 


So-called "stainless" oils have several times been referred to. 
Really stainless oils do not exist; any oil, whether pale or dark 
in color, whether mineral, vegetable, or animal, will in time pro- 
duce a visible stain, but the term stainless as applied to textile 
practice usually means that during the scouring or washing proc- 
ess that most fabrics undergo, oil stains will disappear. 

Oil stains take the form of drops, splashes, or streaks. They 
may be due to oil's dropping from overhead shafting, or oil may 
have got on to the yarn owing to overoiling the top roller bearings 
in the spinning frames. Weavers sometimes cover up defects by 
smearing with dirty oil to escape detection. Oil stains have 
been caused by greasing the reed, but the most frequent cause 
is oil throwing from the cranks operating the sley and from the 
cams actuating the pickers; such splashes show up chiefly oif the 
warp. Stains are also caused by oil splashes from the picker 
spindle in the shuttle box. Hence the reason why a stainless 
picker-spindle oil is nearly always used, even if the loom oil 
employed for other parts* of the loom is not stainless. 

As to oil dropping from overhead shafting, the oil stains pro- 
duced are often difficult or impossible to remove, owing to the 


presence of fine metallic wearings in the oil, chiefly iron. Iron 
stains become red ; copper or brass stains may become black, gray, 
or greenish. 

Mineral oils give a permanent stain on fabrics, and the darker 
the oil the more objectionable the stains. Even bloomless oil or 
oils so pale as to be almost water-white will in time become 
yellow, owing to oxidation, and the color will continue to deepen 
with time. The longer the interval between producing the stain 
and the attempt to remove it (scouring), and the less severe the 
scouring process, the less oil will be removed. If only a short 
time has passed, stains may be removed by dabbing with lard 
oil, olive oil, or other fixed oil, which by blending with the mineral 
oil makes it stainless; i.e., it can be removed by scouring with soda 
lye in the ordinary way. 

Cotton cloths are bleached, and mineral-oil stains are decom- 
posed in this process, by the successive attacks of alkali and chlo- 
rine. For a time after bleaching the oil stains will not appear, 
but after several months the stains begin to show up yellow. 

The best remedy for oil stains is to take precautions that none 
is formed. In many weaving mills, shafting is grease lubricated 
for this reason; or, if oil is used for the bearings, they are well 
fitted up with splash guards and save-alls, which prevent the oil 
from dripping from bearings or creeping along the shafting and 
then dropping. 

When it is considered necessary to have a stainless oil, the 
degree of stainless properties required depends upon the length of 
time the goods are stored before scouring and upon the severity 
of the scouring operation. Speaking generally, an admixture of 
15 per cent of good-quality animal oil or equivalent nondrying 
fixed oil will impart to the spindle or loom oil sufficient stainless 
properties for the majority of conditions. 

In cotton mills many looms require stainless oils only for the 
picker spindle. 

In woolen and worsted mills stainless loom oil should be used 
for lubrication throughout for all looms weaving high-class cloth, 
e.g., dress cloth or such cloth as is used for naval uniforms. 

For low-woolen goods, blankets, etc., stainless oil? are never 


In linen mills stainless oils are not infrequently used for high- 
quality goods, but in jute mills stainless oils are rarely if ever 


called for, as the material is not of sufficient high quality to justify 
the extra cost of stainless oils above the cost of ordinary loom 

In hosiery factories, for material such as woolen underwear and 
light-colored stockings, stainless oils must be used, as the fabric 
invariably gets more or less soiled with oil during manufacture. 
This point is .so important that many hosiery factories when 
testing the oil for stainless properties soak a piece <rf fabric with 
the oil, keep it in stock for a certain time, and then scour it to see 
whether the oil can be entirely removed. 

In lace and curtain factories pure neat's-foot oil is often used, as 
the fabrics receive only a gentle washing, and the oil must scour 
out very easily. Not infrequently the fabrics are not washed at 
all, and it is then absolutely necessary to have an oil as pale and 
as stainless as possible. 

Neat's-foot oil meets the requirements. It is almost colorless, 
and even if there are oil stains on the lace or curtains they will be 
removed the first time that they are washed. 

In many special industries such as corset manufacturing, the 
thread used for stitching is oiled occasionally in order to lubricate 
the needles in the machines. As the corsets are not washed, the 
oil must be as pale and as stainless as possible. Again here, 
neat's-foot oil or a mixture of neat's-foot oil with water-white 
mineral oil is required. If there is a considerable percentage of 
mineral oil in the mixture, the oil stains will in time become 
yellow, so that for white goods which are kept in stock a long 
time this is an important point to keep in mind. 

The table on page 325 gives the author's specifications for 
spindle and loom oils. 

As to the nature of the compound, rape oil has been used with 
success, but it is inclined to gum and tarnish, particularly 
where frames or machinery are exposed to sunlight. With 
blown rape the tendency to gum is still greater; animal oils have 
much less tendency to oxidize and should be preferred; sperjjti 
oil is excellent but very expensive; lard oil or pale whale oil will 
give good results; if desired, they may both be used together in 
the same spindle or loom oil. When stainless properties are 
required (2S, 38, and 4S), a small percentage of olein, say not 
exceeding 3 per cent, is an advantage, as it has good emulsifying 



The mineral base of the oil should be pale in color, but it does 
not matter whether it is an acid-treated or a neutral filtered oil. 

Lather oil (see page 329) must possess exceptionally good stain- 
less properties; it must therefore be made from pale-colored, 
preferably water-white, mineral oil and a large percentage of fixed 
oil, say 30 to 35 per cent, and its free fatty acid contents must not 




at 50C. 

per cent 

Spindle oil 1 . . 




Spindle or loom oil 2 



5 to 6 

Spindle or loom oil 2Sf 



15 to 20 

Spindle or loom oil 3 ... 



5 to 6 

Spindle or loom oil 3Sf 



15 to 20 

Spindle or loom oil 4 



5 to 6 

Spindle or loom oil 4Sf 



15 to 20 

Lather oil 

2 to 3 

8 to 10 

30 to 35 

* See table, p. 57. 

t The letter S indicates stainless properties in the oil. 

exceed 5 per cent; more acid will cause trouble with rusting of 
the needles and other parts. A suitable lather oil may be made 
from 24 per cent rape, 6 per cent pale whale, 3 per cent olein, and 
67 per cent water-white mineral oil of low viscosity, say 75 to 
100 sec. Saybolt at 104F. 

Each factory has its own formula for lather-oil mixture. The 
following is typical : 

Lather oil 3 gal. 

Hard household soap 7 Ib. 

Water . 18 gal. 


Engine Room. Steam engines, chiefly of horizontal construc- 
tion, are used largely for driving textile mills; generally, they 
drive the various mill floors by rope drives from the flywheel. In 
modern mills electric driving is not infrequently used, the genera- 
tors being operated either by steam engines or by turbines, only 
rarely by gas engines. 



As to the lubrication of these engines, the reader is referred to 
the information given under the respective headings. The author 
would mention only the desirability of using compounded steam- 
cylinder oils and using a lower viscosity, preferably filtered 
cylinder oil in the large low-pressure cylinders. 

The practice of using very viscous oil, even cylinder oil, on the 
guides is not a desirable one; an engine oil like bearing oil 4 1 will 
generally be found suitable for external lubrication 
throughout, as well as for the second-motion shaft 
bearings (rope race). When main bearings or 
crankpins are difficult to keep cool with this oil, 
marine-engine oil 1 or 2 (see page 267) may be 
recommended, even with a gravity-circulation 
system, which is frequently employed in textile 

Mill Shafting. The shafting generally operates 
at rather high speeds from 160 to 350 r.p.m. and 
the bearings are either bottle oiled or ring oiled. 
The countershafting, gallow pulleys, etc., are 
often hand oiled. As such hand oiling is a tedious 
occupation, it being difficult to reach the bearings, 
a shafting oiler is often used, as illustrated in Fig. 
128. By pulling the trigger (1), the piston (2) is 
depressed against the action of a spring and dis- 
charges a small amount of oil through the feeding 
tube (3). 

In many mills the engine oil used in the engine 
room is also used for the mill shafting, and the 
waste in power caused hereby is, on the average, 4 
128. per cent of the full mill load. The engine and 
shafting load (transmission load) is approximately 
25 to 30 per cent of the full mill load, and the 
saving in power by introducing bearing oil 2, see page 135, which 
the author recommends generally for mill shafting, is roughly J 5 
per cent of the transmission load. It is a rare thing to find shaft- 
ing oils in use lower in viscosity than bearing oil 3, and against 
this oil bearing oil 2 will save about 10 per cent on the trans- 
mission load, 
i See p. 135. 

shafting oiler. 




Spinning Mills. Frequently one oil is used throughout, 
except for ring spindles, which are always given a separate oil, 
similar in viscosity to spindle oil 2. The mill oils generally used 
have viscosities ranging from viscosity 5 to 6 (see page 57). 
The oils are often straight mineral but are sometimes compounded 
with 5 to 10 per cent of fixed oil. 

The author, however, recommends spindle 
oil 3 for general mill lubrication of preparing 
and spinning departments as well as for 
countershafting and gallow pulleys. 

For ring spindles, spindle oil 1 is recom- 
mended. For high-speed mules, flyers, and 
all cap spindles, spindle oil 2 is recommended 
in preference to spindle oil 3, as it gives an even 
greater reduction in power compared with the 
oils generally employed. 

When the spindle bearings begin to get dry, 
the spindles "whistle," vibrate ("dance 7 '), 
and frequent breakages of the yarn occur. 
With compounded oils the tendency to run dry 
will always be found to be much reduced, as 
compared with straight mineral oils. 

Compounded oil must not be used for ring 
spindles, as in time it will produce a gummy 
deposit which will interfere with lubrication, 
choking the vertical passage in the bearing. If the oil is of 
too low viscosity or badly refined, it will cause continuous wear 
on the step bearing, so that notwithstanding repeated cleaning 
the oil will always become discolored. 

The pump illustrated in Fig. 129 is used for the purpose of 
extracting old oil from bath-spindle bearings before they are 
cleaned and reoiled. The pipe is inserted in the spindle bearing; 
the piston is operated up and down by the handle, drawing the 
dirty oil out from the bearings and discharging it into the main 
barrel of the pump, which can afterward be emptied. 

In wet-flax spinning, a special oil, say spindle oil 48, must be 
used for oiling the flyer spindle tops, when they are of the open 

FIG. 129. 


type. Many mills use lard oil or olive oil, but these oils are 
unnecessarily expensive and no better than the oil just men- 
tioned. The advantage of compounded over straight mineral 
spindle oils is that lower viscosity oils can be used and yet they 
will be found to be more "oily" than the more viscous straight 
mineral oils; therefore they reduce friction and last longer. 
They seem to form a very tenacious oil film in the spindle bear- 
ings and are displaced only with difficulty. This is no doubt 
due to the presence of the fixed oil which we know excels mineral 
oil in the property of oiliness. 

The best practice is to oil the neck bearings of flyer spindles 
and mule spindles while the frames are running. This is Conti- 
nental practice and means that thinner oils can be used efficiently, 
as one is always certain of the necks' being thoroughly oiled. 
Oiling when the spindles are standing, as is the practice in Eng- 
land, may mean that in some necks, particularly when they are 
worn, the oil runs straight through the clearance between neck 
and collar. When neck bearings are fitted with felt pads for 
lubrication, it is immaterial whether they are oiled while the 
spindles are running or when standing. 

As to typical savings in power accomplished on spinning 
frames, see pages 330 to 334. In most mills the oils recommended 
above will save over 8 per cent of the departmental loads, equiv- 
alent to 6 per cent of the full mill load. 

Top Rollers. In some cotton mills spindle oil 3 is often 
sufficiently viscous for the top rollers; but in others, as well as 
in most other spinning mills woolen and worsted mills in par- 
ticular a more viscous oil is required; and, speaking generally, 
the engine oil used in the powerhouse will prove very suitable as a 
top-roller oil. In flax mills a special preparing-room oil is often 
used for the preparing room and for the press rollers in particular. 
This is found necessary when the roller covers are not fitted with 
flannels; but if they are so fitted, an oil like bearing oil 3 can 
be used satisfactorily. Tallow or white tallow greases are qften 
used for top rollers, especially when they are badly worn and 
therefore difficult to lubricate; bad lubrication of the top rollers 
causes a jerky motion and unevenness in the yarn. 

Traveler. The ring upon which the traveler moves in a ring 
spinning frame should be sparingly greased with clean tallow. 


Combers. For combers a viscous oil like the engine oil used 
in the engine room must be used for lubrication of the slow-mov- 
ing parts, whereas the spindle oil should preferably be used for 
the dabbing motions. When the circles are highly steam heated, 
a noncarbonizing very viscous mineral oil is required, having a 
high viscosity about No. 11 (see page 57) and made from a 
pale nonparaffinic-base distilled oil mixed with good-quality 
filtered cylinder stock. 

Weaving Mills. Loom oils should preferably be compounded 
for the same reason as given for spindle oils. 

The percentage of compound need not be more than 6 unless 
particular stainless properties are required. For high-speed light 
looms, loom oil 2 or 3 is recommended ; and for slow-speed heavy 
looms, loom oil 4. The oils must, of course, be sold as loom oils; 
if branded spindle oils, they would almost certainly be condemned 
by the mill people. As stainless loom oils or as stainless picker 
spindle oils (particularly when overpick motion is employed), 
loom oils 2S, 3S, and 4S are recommended. 

Bleaching, Dyeing, Printing, and Finishing. Bearing oils 3 
and 4 are generally used. The calenders, however, require a 
very viscous oil, such as bearing oil 5 or oil even more viscous. 

High-melting-point greases, with melting points suitable for 
the temperature of the bearing journals, are also used. 


Hosiery machines are chiefly knitting machines and either 
straight-bar machines, knitting flat pieces of material; or circular 
machines, knitting tubular pieces. Bar machines have several 
hundreds, and the largest circular machines many thousands of 
needles. The needles require some slight lubrication so that 
the yarn may pass easily through them. The lubrication is done 
by the yarn, which before entering the machines is passed through 
a trough containing emulsified-lather oil; as the yarn leaves the 
trough, surplus lather is squeezed out by rollers. 

For general lubrication of most circular machines and power 
sewing machines, loom oil 2S will be found suitable. Bar 
machines require a somewhat heavier oil, such as loom oil 38, 
and this oil may also be recommended for circular machines that 
have become worn. 


Stainless properties are practically always required, and the 
oil ought to be thoroughly tested in this respect, as mentioned on 
page 323. 


Very great reductions in power can be accomplished by pay- 
ing careful attention to the selection of suitable oils for each 
department in the mill, as well as for the mill shafting and the 

In a ring-spindle frame, for example, about 80 per cent of the 
power is required for driving the frame empty, only 20 per cent 
being consumed in handling the yarn. In the case of preparing 
machinery, an even greater percentage of the full-load power is 
required to run the machines or frames empty. 

In a jute-spinning frame of the ordinary type the power con- 
sumed usefully is very much the same as in a ring-spindle frame, 
but in the modern spinning frames, in which the tension of the 
driving bands is kept uniform, there is a great reduction in the 
power consumed by the frame, and only 65 per cent of the full- 
load power is required for running empty. 

In mules or looms a great portion of the power is used in over- 
coming the inertia of the moving parts which have to be acceler- 
ated, stopped, and, in the case of the loom, quickly changed. In 
the loom, for example, the sley moves backward and forward 
quickly; the picker motion just as quickly; and the shuttle is 
thrown quickly to and fro, all of which requires a great deal of 
power, so that the percentage of power influenced by lubrication 
in a mule or loom is less than in ring-spinning frames. 

In the average steam-engine-driven textile spinning mill, 1J^ 
to 2 Ib. of coal is consumed per indicated horsepower per hour; 
and the heat value actually converted into useful work in the 
form of preparing or spinning the yarn, etc., will not be more 
than 1J^ to 2 per cent of the heat value of the coal used unde* 
the boilers. 

The possible saving in power by introducing correct grades of 
spindle and loom oils is nearly always considerable. To take an 
example: On a ring-spinning frame using an oil like spindle oil 
3, another oil like spindle oil 1 was introduced. The results 
were as in the table on page 331. 



The saving in power in this case amounted to 8.8 per cent 
and indicates the results that can be obtained in most textile 
mills, as the first oil used is typical of the ring-spindle oils now 
in general use and is quite unnecessarily viscous, except perhaps 
for frames with old and worn Rabbeth spindles. Whenever a 
change from a viscous to a less viscous oil is carried out, the 
low- viscosity oil will turn black, the discoloration being due to 
extremely fine metallic particles from the rubbing surface. In 
other words, very slight wear takes place, the surfaces adapting 
themselves to the new oil. After the pumping-out and recharg- 
ing process, the fresh oil should work perfectly clean. 

Such a saving in power is worth many times the value of the 
oil itself, and, in addition, the yarn produced by the frame will 
be found more uniform, because of the smoother running of the 


Number of spindles 300 

Diameter of line-shaft pulley 40 in. 

Diameter of frame pulley 15 in. 

Diameter of tin roller 10 in. 

Diameter of whorl 1 in. 



1. Influencing conditions: 
Counts spun 



Weight of yard per doff, pounds 


14 6 

Room temperature, degrees Fahrenheit 



Relative humidity, per cent 



2. Power (measured by Emersons dynamometer) : 
Brake horsepower 



3. Temperatures, degrees Fahrenheit: 
Temperature of spindle rail .... 



Frictional heat 



4. Loss due to belt and band slip : 
Speed of line shaft, r.p.m . . . 



Theoretical speed of tin roller 



Registered speed of tin roller 



Belt slip, per cent 



Theoretical speed of spindles 



Registered speed of spindles 



Driving-band slip, per cent 


4 9 



Improved lubrication means lower frictional heat, which is 
evidenced by a lower rise in temperature of the spindle rail 
above the room temperature. 

The driving bands which run over the tin roller and drive the 
spindle always slip slightly; when they are in proper condition 
the slip should not be more than a few per cent. The lower 
friction of the spindles will reduce the band slip and thus slightly 
increase the spindle speed, as shown by the test.' Less band slip 
also means less wear of the driving bands, and the annual con- 
sumption of driving bands is quite a good indication of the quality 
of ring-spindle oil used. The reduced power consumption of the 
frame will tend to decrease the belt slip in the driving belt, and 
this effect is also shown in the test figures. 

In a worsted spinning mill a test was carried out on a spinning 
frame having 216 open-type flyer spindles. The oils in use on the 
two tests were oils A and B. Oil A was a straight mineral oil 
having a viscosity of 5 centipoises at 50C. Oil B is spindle oil 2, 
specified on page 325. The power measurements were recorded by 
an Emerson dynamometer, and, besides particulars of the horse- 
power, readings were obtained of the rail temperature, room tem- 
perature, relative humidity, tin-roller speeds, and spindle speeds 

Oil in use 

required to 
drive frame 

Rise in 
of spindle rail 







Reduction in horsepower required to drive frame. . .0.24, or 10.0 per cent 
Reduction in temperature of spindle frame 4.9F., or 47.6 per cent 

Oil in use 

Tin-roller speeds per minute 

Spindle speeds per mini^e 



per cent 



per cent 









every 10 min. for 2 hr. in the forenoon and 2 hr. in the afternoon. 
Before testing, the frame was cleaned and well oiled with oil A ; 
and after the first test was completed the footsteps were again 
wiped out, and oil B was put into use. The frame was then 
allowed to run for a full day before the second test took place. 

The temperature of the atmosphere and the relative humidity 
were the same on both tests. 

In one mill the introduction of spindle oil 2 for cap spindles 
reduced the wear of the driving bands very considerably. It 
was brought to the overseer's notice that the band boy had very 
little work to do; when the boy was asked why he did not attend 
to the bands, he replied that as soon as the new oil was put into 
use they seldom broke, and he had none to repair. 

In another case the power consumption of the frames with 
oil A was so great that the belts were always slipping on the 
pulleys. It was not possible to get all the spinning frames run- 
ning until seven A.M., as it took considerable time before 
the oil became warm and fluid enough to reduce the power con- 
sumption of the frames. The steam engine driving the mill was 
hardly powerful enough to cope with the load. 

Comparative power tests in textile mills should therefore 
never be carried out on Mondays when the mills have been shut 
down for the week end. The oil cools down in the bearings, and 
the starting load on the Monday morning is always considerably 
higher than later on during the week. 

When engines are overloaded, it is often difficult to start the 
mill on full load on Monday mornings in particular and it may 
even be necessary to leave out one or two departments until the 
engine eventually is able to cope with the load. The introduction 
of more suitable grades of oil reduces the horsepower required 
and particularly the starting horsepower in the early part of 
the week, so that the engines are able to get up their normal speed 
much more rapidly and maintain their speed more uniformly 
during the day. There have been many cases of overloaded 
engines which after a change in lubrication have been found quite 
powerful enough to drive the mills, so that a study of the lubri- 
cating conditions has saved such mills the heavy expense of put- 
ting in a new engine or introducing electric motors to take care 
of part of the load. When better lubricants are introduced, the 
improved working of the machines is soon observed by the easier 


starting of the machines or by their running for a longer time 
after the driving belt has been moved on to the loose pulley. 

Quite a simple test for the engine and shafting load is to run 
the engine and shafting at the lunch hour when all the machinery 
in the mill is stopped; then shut off steam and observe the 
number of revolutions made by the engine before it comes to a 
standstill and the time taken. An improvement in lubrication 
is immediately shown by the greater number of revolutions and 
the longer time that passes before the engine comes to rest. 

When an appreciable reduction in power has been accom- 
plished in a textile mill by introduction of better lubricants, the 
main effects are the following: 

1. A reduction in the total horsepower of the mill as well as in the engine 
and shafting load and the power consumed by each department in the mill. 

2. A reduction in the amount of coal required for power purposes. When 
the reduction in power is appreciable it should always be possible to find a 
corresponding reduction in the coal consumption, particularly when the 
amount of coal used for heating and power are kept separate. 

3. A reduction in the temperature of all bearings and spindle bases. 

4. An increase in the speed of countershafting, machines, and spindles 
due to reduced slipping of driving belts and driving bands. If the engine has 
been overloaded, the reduction in power will bring about an increase in the 
engine speed, and the engine will reach its normal speed more quickly after 

5. A slight increase in production, due chiefly to fewer stoppages, as many 
stoppages are caused through defective lubrication. 

6. A decrease in the wear and tear of the machines as well as of belts 
and driving bands. The decrease in the wear of the driving bands may 
often be quite considerable. 



The tubs in collieries are known by many names, such as trams, 
hutches (Scotland), and mine cars (United States). The follow- 
ing remarks apply chiefly to mine cars in collieries. 

Their lubrication consumes on an average 50 per cent of the 
oil used in a colliery and is of great importance, as trouble with 
the tub lubrication may easily cause reduced output. The tubs 
are preferably made of steel; with wooden tubs dust shakes 
through the floors, contaminates the axles, and interferes with 
lubrication. Their carrying capacity is from 4 cwt. to 2 tons. 
Tubs have two axles usually of rolled steel ranging in diameter 


FIG. 130. Open-type bearing. 

from \Y to 2 in., wheels ranging from 7 to 16 in. in diameter, 
and bearings of a length preferably not less than twice the 
diameter of the axle. 

Wheels and Axles. With fast wheels the wheels are riveted to 
the axle, which revolves in the cod bearings. 

With loose wheels both wheels are loose on the axle, which does 
not revolve. 

With loose wheels and axles the wheels as well as the axles are 
free to rotate. Where the track has many curves this system or 
a combination of one fast and one loose wheel is often used. 

Cod Bearings. These may be either outside or inside bearings 
and either open or enclosed. Figure 130 shows a typir al open- 
type bearing. The spectacle plate (1) should be bent well to one 
side, so that it does not foul the automatic oilers when the tub 
passes over them. The bolts (2) should preferably be put in 
from the bottom and must not project below the axle, as in Fig. 




131, when they will foul the oilers. Figure 132 also shows an 
undesirable condition from the oiling point of view, and it may 
be produced by excessive wear of the bearing shown in Fig. 130. 
Cod bearings may be of cast iron but are usually of cast steel, 
and where there is no dust or the dust is not of a gritty nature they 

Fio. 131. 

Cod bearings. 

FIG. 132. 

are preferably lined with white metal. The question of grit is 
of importance only when the speed of the tubs is sufficiently great 
to raise the dust to any extent. 

When, as is sometimes the case, the bearing entirely encloses 
the axle, or the spectacle plate is in the center, the axles cannot be 

FIG. 133. Rowbotham wheel. % 

oiled automatically, except by squirt oilers, e.g., the Abbott oiler, 
which will be referred to later. Figure 133 illustrates the Row- 
botham wheel, much used in South Wales on account of the fine 
dust existing in the mines. The wheel is loose on the axle, and 
the hub serves as an oil reservoir; the oil is injected by a syringe 
against a self-closing ball valve, as shown. Enclosed wheels 



of similar types are much used in the United States, frequently 
employing oil-soaked waste; and on the Continent there are 
several types of roller or plain bearings so arranged that the 
bearing housings at either end of the axle are combined and form 
a sleeve surrounding the axle, as shown in Fig. 134. The space 
between the axle and the sleeve is filled with oil through a filling 
hole in the center. The lubrication is very economical; one filling 
may last a month or more. The oil works its way out through 
the ends and keeps the bearings clean. The latter should have 
good felt packings when using oil, as otherwise it works out too 
freely and is wasted. With roller bearings a very soft grease is 
better than oil and more economical. 


Hand Oiling. In some mines the tubs are still oiled by hand, 
although this practice is fast disappearing. The tubs are turned 

FIG. 134. Roller bearings for mine cars. 

over first on one side, then on the other; the oil is applied through 
"coffeepots;" loose wheels are given a "spin" when being oiled 
to get the oil well worked into the bearing. By flattening the end 
of the oilcan spout it is possible to reduce the waste of oil to 
some extent. It is good practice to use thick oil and heat it in a 
steam-heated tank; cold oil will not run through a flattened spout; 
and if an ordinary wide spout is used, most of the oil will be 



With fast wheels, the axles may be oiled by hand by means of 
a brush; to avoid undue waste, the brush should not be dipped 
in the oil but into cotton or wool waste kept well soaked with it. 

Hand greasing may be done by a stick or a brush but is always 
very wasteful ; the surplus grease drops on to the track and makes 
it greasy and dirty. 

Mechanical Oilers. Figure 135 shows an early type of greaser 
a scalloped wheel connected to the axle by a spiral spring, 
which allows the wheel to be depressed when the axle passes over 
it and gets smeared with grease. Coal dust gets into the trough 
and gives trouble. When the grease is thick or becomes thick, 
owing to cold, the wheel cuts a track in it and revolves without 

FIG. 135. Scalloped- wheel greaser. FIG. 136. "Knockout" greaser. 

lifting the grease. Revolving brushes have been used instead of 
the wheel but are very wasteful indeed. 

The " knockout" greaser (Fig. 136) is a better form of greaser; 
it is simple and accessible; the wheel can be lifted right out. This 
greaser is also used for oil but should then have a brake fitted so 
that it soon stops after the axles have passed ; otherwise it causes 
waste by throwing the oil. 

The disadvantage of this and similar greasers when using oil 
is that the oil drains off during an interval, so that when the next 
set of tubs comes over, the first few tubs do not get properly oiled. 
All tub oilers should be so designed that none of the axles can pass 
over without being oiled. Most types have some form of pump 
actuated by the tub axles; the wheels of the tubs passing overman 
oiler should therefore be of the same size and as uniform as pos- 
sible. When the wheels are much worn, the axles of those par- 
ticular tubs are nearer the ground, depress the pump plunger too 
much, and cause waste of oil. 

The oiler or its foundation should be secured firmly to the rails; 
otherwise, it may be pushed into the ground with the result that 
the axles no longer depress the plungers sufficiently. 



Figure 137 illustrates an oiler designed by W. A. E. Woodman 
and the author. It is suitable only for open-type bearings (Fig. 
130) and fast wheels. It has a large container; the lid carries 
the pump barrel with its suction valve. The bow is guided by 
two vertical guide bars and carries the pump plunger with a 

delivery valve at the bottom. The guide bars are connected by 
a crosspiece, which is forced upward by a spring against a stop; 
the stop is adjusted vertically by a single outside adjustment, 
thus determining the depression of the plunger when the axles 
pass over the bow, and wipes the oil from the oil delivery well in 
the center of the bow. The lid entirely covers the container 
and is provided with a large filling hole; the edges of the hole are 


raised above the level of the lid, to prevent dust and dirt from 
getting in when the container is being filled. This is a very 
important point, and for the same reason the filling lid is so 
designed that it cannot be left open but automatically falls and 
closes the opening. 

In many types of oilers for open-type bearings, the haulage 
ropes and coupling chains are liable to get underneath the bow and 
bodily pull the oiler out of the track. This has been provided 
against by placing at either end of the bow a fin, which is cast on 
to the lid and gives an inclined plane for the rope or chain to run 
up and slide clear over the bow. These fins also act as buffers 
against severe end shocks. 

The oiler has to be well made, but in the author's experience 
tub oilers cannot be made too well. The oiler shown has worked 
under very severe conditions in South Wales (very heavy tubs) 
where no other oiler has been able to stand up to the conditions. 
After many months' working, no perceptible wear had taken 
place, no dirt had got into the container (only a coarse sieve 
is provided), and the adjustments had never been touched. 
Equally good results have been obtained in Lancashire and other 
collieries, where the conditions are much less severe. 

The oilers are usually placed on the same foundation, but 
sometimes it is best to stagger them. This gives more room 
for the ponies (where ponies are used) and is also advantageous 
where the spectacle plates are inclined to foul the oilers. The 
rail at the first of a pair of oilers and a little before is raised, say, 
1^3 in. above the other rail. This makes the tub body slide 
over toward the lower rail and gives more clearance to oil the 
underside of the axle in the cod bearing. The same performance 
is reversed at the next oiler which is placed, say, 10 to 15 yd. 
farther along the track. 

On the surface the oilers should not be exposed to rain but 
placed under a roof or shelter. Down pit the oilers should be 
laid down in a dry place and not where surface or roof water*is 
likely to come in contact with them. On entering a wet district, 
the tubs should be oiled so that the oil film will last until they 
reach the next oiler, which should be placed immediately after 
the wet district. It is much easier to renew and maintain the 
oil film if it is never allowed to be completely washed away; the 
oil does not adhere well to a wet axle. 



Figure 138 shows one form of the Abbott oiler oiling outside 
axle bearings of the type shown in Fig. 132, which can obviously 
not be oiled by the oiler shown in Fig. 137. The plungers (1) 
are depressed quickly and so timed that they squirt oil on to the 
underside of the axles; surplus oil is caught by the save-all (2). 

FIG. 138. Abbott oiler. 

The oil in Abbott oilers must be steam heated to give a good 
and uniform squirt; for this reason they cannot be used down pit, 
where there is no steam. 

Distance between Oiling s. Apart from wet districts, it may be 
raid that the larger the wheels and the cooler the pit the longer 
the distances that can be allowed between the oilers, other things 
being equal. The axles should never be allowed to run dry; it is 
better to space the oilers closer together and give them less oil per 


oiler than to space them so far apart that there is a risk of under- 
lubricating the axles. Efficient oiling saves wear and tear and 
means less power required for hauling the tubs. It is good prac- 
tice not to exceed 1^ miles between oilings, and with small 
wheels and narrow bearings oiling every mile will be required. 

In some mines using a very viscous oil, tubs have run as 
much as 4 miles on one oiling, but the lubrication has not been 
all that could be desired. 

Before grades of tub oils and greases are described, it is 
necessary to refer to the various systems of haulage. 


With endless-rope haulage the empty tubs are continuously 
and slowly hauled into the mine and return loaded, the speed of 
haulage being from 2 to 4 m.p.h. 

With main haulage the shaft is inclined toward the workings, 
the incline exceeding 1 : 24. The empty tubs run into the shaft 
by gravity and are hauled out loaded; only one rope and one 
haulage drum are employed. 

With main and tail haulage two ropes are used; the main rope 
hauls the loaded tubs out, and the tail rope pulls the empty ones 
in. The haulage drums are both operated by the same engine. 
The speed for main and tail haulage may be as high as 20 m.p.h. 

The haulage ropes are driven either by a steam haulage engine 
or by an electric haulage engine. In the United States ropes 
are discarded in many mines, and the mine cars pulled out or in 
by electric or compressed-air locomotives. 


Grease can be used only for slow-speed conditions; if used on 
main and tail haulage, it gives a great deal of trouble and wastes 
much power in hauling the tubs; also, the wear is excessive. 
Good tub oil as compared with grease gives cleaner and better 
lubrication, and not only does it save a great deal of power, but 
rightly applied it also saves in cost. In most cases where a 
change has been made from grease to oil, the saving in consump- 
tion is 50 per cent or over. With electrically operated haulage 
the difference between grease and oil or even different qualities 
of oil is readily observed. 

With ball or roller bearings there is very little difference in 
friction between oil and grease. Very little lubricant is required, 


but it must be of good quality (see under Ball and "Roller 

The lubricant must be selected according to the temperature 
of the mine, whether it is dry or wet, the amount and nature of 
the dust, the type of oiler employed, and the distances between 

Temperature. In deep, badly ventilated pits the temperature 
is higher than in shallow well-ventilated pits. The higher the 
temperature the more viscous the oils required. In cold pits 
a good cold test will be required, so that the oil will not be too 
sluggish in the automatic oilers. 

Wet Pits. In very wet pits good-quality grease may have to be 
used, if the tubs must run long distances between oilings. It 
is not so easily washed off the axles as is oil. Tub oils for wet 
pits should have a tendency to emulsify with water. 

Dust. Fire-clay dust and coal dust have a drying effect on the 
oil film. Free-flowing oils and frequent oilings are desirable with 
these dusts. 

Stone or flint dust will cause heavy wear; the wear can be 
minimized only by using viscous oils of good quality and by 
frequent oilings. If a sticky oil is used, the dust forms a grinding 
paste with the oil and causes heavy friction and wear. 

Type of Oiler. Wheel greasers (Figs. 135 and 136) can use oil 
of almost any description and also grease, as long as it is not 
too thick. 

Abbott oilers can use oils with a poor cold test, as they are 
usually steam heated. Such oils solidify on the cold axles and 
form a film with good lasting properties if the oil is of good 

Pump oilers (Figs. 137 and 138) when not steam heated cannot 
use oils that are so sluggish at the working temperature that the 
pump fails to act. 

Hand Oiling. When the oil is heated, it need not have a good 
cold test, but this may be necessary when it is not heated. 

Distance between Oilings. With long distances between oil- 
ing more viscous oils are required than when the oilers are closer 


It is customary to use black oils one for summer use with a 
cold test of 25 to 30F., and a winter oil with a cold test of 5 to 



15F. These oils are dark residual oils from distilling lubricating 
crudes or from redistillation of lubricating-oil distillates, or they 
are mixtures of such oils with low-viscosity, low cold-test oils, so 
as to produce oils of the right viscosity and cold test. 

The asphalt contents should preferably not exceed 3 per cent 
in the better quality oils; but for rough service, oils with much 
higher asphalt contents have been used. Typical viscosity 
figures for black oils are given in the table below. 

Black oil 

number 1 

at 50C. 

Cold test, 




M 5 




2 %o 




5 %o 

i See table, p. 57. 

Blaek tub greases are usually rosin greases. Sometimes 
so-called " floating greases/' containing talc, are used. 

There are various formulas, and the better qualities contain no 
filler. A rough test for the presence of filling material is to burn 
a sample and examine the residue (lime, talc, etc.). 



Stationary and Marine Engines. 
With special sections on: 

Corliss Value Engines. 
Colliery Winding Engines. 
Uniflow Engines (Stumpf Engines). 
Marine Engines. 



Steam engines are the most reliable and most highly developed 
and specialized of all power producers. 

Most land steam engines are horizontal, and practically all 
marine engines are vertical, except a few "inclined" engines 
employed in paddle steamers. 

They can be classified according to: 

Arrangement and number of cylinders. 
Type of valves employed. 

Arrangement and Number of Cylinders. Steam engines may 
have one, two, or three cylinders side by side, all using high- 
pressure steam. 

Two-cylinder engines twin engines, mostly horizontal are 
used as colliery winding and haulage engines or steelworks 
rolling-mill engines. 

Three-cylinder engines triple engines, mostly horizontal are 
used as steelworks rolling-mill engines. 

Engines in which the steam expands in two, three, or four 
consecutive stages are called compound, triple-expansion, and 
quadruple-expansion engines, respectively. Some triple-expan- 
sion engines have two low-pressure cylinders and are therefore 
fourrcylinder, triple-expansion engines. 



There is a tendency to turn away from triple-expansion engines 
in favor of compound engines of various types, such as four- 
cylinder compound engines with two high- and two low-pressure 
cylinders or three-cylinder compound engines with two high- and 
one low-pressure cylinders or one high- and two low-pressure 

The two cylinders of a compound steam engine may be 
arranged one behind the other a tandem engine or side by 
side a cross-compound engine or with horizontal, high-pres- 
sure and vertical, low-pressure cylinders an angle-compound 

Types of Valves. Many types of valves are in use, but they 
may be divided into four main groups, as follows: 

1. Slide valves. 

2. Corliss valves. 

3. Piston valves. 

4. Drop valves, or poppet valves. 

Slide valves are not used in single-cylinder engines of over 125 
hp., because they are inefficient. 

In compound and triple-expansion engines, slide valves may be 
used for the intermediate- and low-pressure cylinders in sizes 
from 50 to 750 hp. per cylinder. 

Slide valves can be used only for low superheat, as their unsym- 
metrical shape causes warping. They can, however, be used at 
high speedy as they are positively operated. 

Corliss valves are used rarely in engines below 125 hp. in size, 
as they are not so adaptable to the high speeds at which small 
engines operate. They can be employed with moderate superheat. 

Piston valves, notwithstanding their rather low efficiency, are 
used even for very large power units, as they can be operated at 
high speed, with high steam pressure and high steam tempera- 
ture, and are very reliable for severe service, as in colliery wind- 
ing engines and steelworks rolling-mill engines. They are used 
largely in marine engines and for locomotives. 

Drop valves, or poppet valves, are used for the highest powers, 
on account of their great efficiency; they are not used for power 
units below 125 hp., for the same reason that Corliss valves are 
not employed. Drop valves can be operated with high steam 
pressure and high steam temperature at higher speeds than the 
Corliss valve but not at such high speeds as the piston valve. 



Land Engines. Below is shown, for the different types of 
valves, the normal range of steam pressure, maximum steam 


Number of cylinders 


Q'f oo YYl 

f ^rv"i 






per cylinder 






60 to 135 


350 to 60 

Up to 125 



Corliss . . . 

80 to 160 


150 to 60 

125 to 2,000 

1, 2, or 4 

1, 2, or 3 

Piston .... 

90 to 


500 to 90 

Up to 3, 000 

1, 2, or 3 

1, 2, or 3 


Drop or 

poppet . 

120 to 450 


180 to 90 

125 to 3, 000 

1 or 2 

temperature permissible, revolutions per minute, horsepower 
per cylinder, and number of cylinders employed in horizontal as 
well as vertical land engines. 

The table on page 348 shows the most frequent combinations of 
valves employed in single-cylinder, twin, triple, compound, and 
triple-expansion engines as used for land purposes. 

Marine Engines. Small marine engines are compound, say 
below 100 hp. for single units. The vast majority are, however, 
triple expansion. Single units above 3,000 i.hp. are frequently 
triple-expansion four-crank engines, with one high-, one inter- 
mediate-, and two low-pressure cylinders. 

Single units above 4,000 i.hp. are frequently quadruple-expan- 
sion engines, with one high-pressure, one first-intermediate, one 
second-intermediate, and one low-pressure cylinder. 

The valves belonging to the high-pressure cylinder are practi- 
cally always piston valves. Piston valves are also generally 
used for the intermediate-pressure cylinder, but sometimes slide 
> valves are used. Slide valves are generally used for the low-pres- 
sure cylinder. 

Practically all marine steam engines are of the inverted verti- 
cal type; only a few have the cylinders and valves lying at an 
angle, as is the case with some paddle steamers. 



Quite frequently the exhaust steam from marine steam engines 
is utilized in an exhaust-steam turbine which transmits its power 
to the main crankshaft by means of gears or chains. 

Steam. In the vast majority of cases, saturated steam is 
employed; but during recent years, superheated steam has come 

Type of engine 



One low- 

Two low- 

Single cylinder 

j Corliss 
j Piston 
! Slide 

j Slide 
j Piston 

1 Corliss 
} Corliss 
V Piston 

engines, { 


winding am 
steelworks re 


i haulage 



Two high-pressure cylin- 
ders side by side 

Three high-pressure cyl- 
inders side by side .... 


Triple expansion 


Three cylinders 

Four cylinders 

into use very largely on the Continent, the maximum steam 
temperature at the engine stop valve being 650F. 

The revolutions per minute of marine steam engines are largely 
governed by considerations affecting the propeller efficiency and, 
therefore, do not vary much for engines above, say, 1,000 hp., 
being generally between 80 and 90 r.p.m. 


In the case of launches, higher speeds are frequently used and 
with consequent lower propeller efficiency. Some large naval 
ships have been constructed with high-speed short-stroke engines, 
the maximum speed seldom, however, exceeding 130 r.p.m. 

The various types of steam engines having now been classified, 
the subject will be treated under the following headings: 


Oil in Exhaust Steam and Feed Water. 

Oil in Boilers. 

Methods of Lubrication. 




Lubrication of Corliss-valve Engines. 

Lubrication of Colliery Winding Engines. 

Uniflow Steam Engines (Stumpf Engines). 

Marine Steam Engines. 

Cylinder-oil Consumption. 

Selection of Oil. 

Testing Cylinder Oil. 

Physical and Chemical Tests. 

Use of Tallow Mixtures and Semisolid Greases as Cylinder Lubricants. 

Lubrication Chart. 


Locomotive-Cylinder Oils. 


The range of steam pressure employed for different engines is 
given in the table on page 347. 

Dry or Wet Saturated Steam. When the steam leaves the 
boiler in a dry condition, it is called "dry saturated steam "; but 
under certain conditions, e.g., when the boiler is forced above 
its normal capacity, or if the water level in the boiler is too high, 
the water boils violently; priming takes place, and the spray or 
foam from the water surface goes out with the steam, which in 
this condition is called "wet saturated steam." 

It is in order to prevent the bulk of this water from being 
carried over with the steam that various so-called "antipriming 
devices " are frequently employed. If the steam pipe is long or 
not properly covered, a fair amount of steam will be cooled and 
condensed into water which is carried along with the steam 
toward the steam engine, together with any water that may have 
been carried over from the boiler. Therefore, the steam pipe 
should be covered with insulating material, to minimize conden- 


sation. Water in the steam should be taken out, as far as pos- 
sible, by a steam separator. But where the steam is very wet, 
it is difficult even with a good separator to prevent some of the 
water from entering the steam engine. 

Superheated Steam. Saturated steam, in passing through the 
heated superheater tubes, is heated above its saturated-steam 
temperature and becomes superheated steam. 

The water that has been carried over from the boiler during 
periods of priming contains impurities, either solid impurities 
or salts in solution. When priming ceases, this water evaporates 
in the superheater, and the impurities will accumulate in the 
superheater tubes, in the form of a dry dust, which gets blown 
over with the steam into the engine and interferes with lubrication. 

The steam separator tends to remove not only water but also 
rusty scale and impurities which are carried over from the boiler 
or which break loose from the inside of the steam pipes, also fine 
oxidized scale from the inside of the superheater tubes which gets 
carried over in the form of a fine black dust. 

Cutting and scoring of cylinders, valves, and valve faces is 
sometimes experienced shortly after starting up a new engine. 
It is seldom due to lack of lubrication or to the quality of the 
cylinder oil used, but in most cases it can be accounted for by the 
steam line's not being properly blown through and cleansed from 
scale, foundry sand, rust, and the like. It is obvious that the 
entrance of such impurities into the steam engine will cause 
trouble, and the utmost care should be taken, when starting new 
steam engines, that the pipe lines from the boilers to the engines, 
as well as the internal spaces in the valve chests, cylinders, and 
steam connections between the cylinders, are thoroughly cleansed. 

The importance of having a steam trap just before the inlet 
for the steam into the engine is not sufficiently appreciated by 
most steam users. If no steam separator be fitted, it is obvious 
that solid matters from the steam line or the boilers will have free 
access to the engine, which often results in the necessity for 
early repairing of cylinders and refacing of valves, etc. * 


A portion of the oil used for lubricating the steam-engine cylin- 
ders and valves will pass out through the valve- and piston-rod 
glands, but the greatest portion will leave the engine with the 


exhaust steam and will be present in the form of oil in suspension 
or oil in emulsion. 

Oil in suspension consists of oil globules which are fairly easily 
removed from the steam by the exhaust-steam oil separator. 
The globules of oil that are not extracted from the steam in the 
separator will, in the case of condensing engines, mix with the 
condensed steam and reach the hot well, where the greater 
portion will rise to the surface in the form of "float" oil which 
can be skimmed off. 

A final safeguard may be provided in the form of a feed-water 
filter, the filter medium being cloth, sand, wood wool, etc., which 
will retain the globules of oil in suspension. The filter gradu- 
ally becomes fouled with the oil, and the difference in pressure 
of the feed water on either side becomes greater and greater. If 
the fouling of the filter is allowed to proceed too far, the danger 
arises that the collected matter may be swept through and carried 
into the boilers. If a pressure gauge is fitted, it shows the differ- 
ence in pressure before and after the filter; and the engineer will, 
from experience, soon become acquainted with the maximum 
difference in pressure permissible. 

In marine practice the danger of the pressure's becoming too 
high is particularly great where the feed-water pump is directly 
driven by the main engines, as, in event of the engine's racing, 
the increased speed of the feed water will certainly tend to clear 
out the oil from the filter and carry it straight into the boilers. 

Asbestos fiber is said to be capable of almost entirely breaking 
up the emulsified particles of oil and water and of thus extract- 
ing the greater portion of even emulsified oil, but, so far, experi- 
ments with such filtering material have not led to any practical 
solution of this question, as asbestos fiber is both costly to renew 
and costly to clean. 

Oil in emulsion consists of minute particles of water (less than 
1/50,000 in. in diameter) coated with an oil film. They are so 
fine that they float in the steam, and consequently the exhaust- 
steam oil separator will remove only a portion of the emulsified 
oil. The greater portion of the oil in emulsion, therefore, mixes 
with the condensed steam, which assumes a milky appearance. 
The greater the amount of oil the more milky will the water be. 

Whereas oil in suspension is fairly easily removed in the hot 
well or in the feed-water filters, not so with the oil in emulsion. 


The particles are so small that they will not rise to the surface in 
the hot well, and the filtering medium in the feed-water filter will 
not be able to retain them. 

Exhaust Steam. In noncondensing steam engines the steam 
passes out into the atmosphere, or it may be used for the purpose 
of heating the premises, for drying purposes, or for heating the 
feed water in feed-water heaters. The presence of oil in the heat- 
ing or drying apparatus reduces its heating capacity considerably. 

In condensing engines the steam, when leaving the engine, is 
condensed either by the jet- or the surface-condensing system. 

Jet-condensing System. The exhaust steam on entering the jet 
condenser meets numerous jets of cold water. The cold water 
condenses the steam into warm water, which by means of a pump 
is taken from the condensing chamber and delivered into the 
hot well, from which a small portion of the water is taken away 
by the boiler-feed pump for boiler-feed purposes. The bulk of 
the water, however, either is allowed to waste or, where only a 
limited supply of cooling water is available, it is passed through a 
cooling tower, where it is cooled, so that it can be used over and 
over again. A large portion of the cylinder oil will separate out 
and present itself as float oil on the surface, which can be skimmed 
off from time to time. There is generally very little chance of 
any cylinder oil's reaching the boilers where the jet-condensing 
system is employed. 

Feed-water Heaters. Such heaters are installed in numerous 
plants ashore, and generally they are what are termed " contact 
feed-water heaters" in which the steam comes in direct contact 
with the feed water. 

Some oils form an emulsion with the water and do not separate 
out in the heater. The greater the quantity of water contained in 
the heater the easier it will be for the oil in suspension to separate ; 
but to get the separation anywhere near satisfactory, it is neces- 
sary to use a pure mineral cylinder oil. Under no conditions 
should the oil be allowed to accumulate in large quantities in 
the heater, but it should be drained or skimmed off at suitable 

It is also good practice to take the suction of the feed pump 
from a point as far below the surface as possible, because near the 
bottom the water is generally most free from oil. Limy deposits 


which accumulate in the bottom of the heater should never be 
allowed to reach the level of the suction pipe. 

Surface-condensing Plant. The exhaust steam is here not 
cooled by direct contact with the cooling water but simply passes 
through the condenser chamber, in which are a great number of 
tubes through which cold water is forced. The steam is cooled, 
condensed, and pumped into the hot well, whence the boiler-feed 
pump takes the whole of this water and delivers it back into the 
boiler, where it is converted into steam and starts the circuit 

All the oil contained in the exhaust steam will accumulate in 
the hot well, and the same remarks that were made in refer- 
ence to contact feed-water heaters apply here as to skimming off 
the float oil and taking the feed water from a low level in the well. 
Also, here oil in emulsion will be carried through with the feed 
water and will enter the boiler unless eliminated by special 
means. Surface-condensing engines in land service are compara- 
tively few in number but are used to some extent for electric- 
power installations and the like and, more especially, where town 
water is expensive; also for ice-manufacturing plants, when the 
condensed water is afterward used for ice production and where 
any trace of oil would make the ice cloudy. 

Example 15. A horizontal, 200-hp., cross-compound, con- 
densing Robey engine was using a common black cylinder oil. 
The engine was surface condensing, and the feed pump dis- 
charged the water through a filter which was supposed to clear 
the feed water from oil. After passing this filter the feed water 
went direct to the boilers. The consumption of cylinder oil was 
found to be 4 or 5 drops per minute; and if the feed was reduced, 
the engine started groaning and grinding. In spite of all that, 
the makers of the filter claimed, oil was found in quantities in the 
boilers and was the cause of a most serious complaint from the 
insurance companies. 

An examination of the boiler deposit showed that it was com- 
posed of black greasy matters and boiler scale. The boiler scale 
was partly carbonates, sulphates, and hydrates of lime and mag- 
nesium. After introducing a pure, mineral, filtered cylinder oil 
it was found possible to reduce the oil consumption to one drop 
in 70 sec., and it was reported that a marked improvement in 



the boiler conditions took place at once, practically all the oil 
separating out in the hot well. 

Extracting the Oil. The oil may be separated from the exhaust 
steam by oil separators or extracted from the feed water by 
chemical or electrical treatment. 

FIG. 139. Baker oil separator. 


Exhaust-steam Oil Separators. Some exhaust-steam oil separa- 
tors are very large reservoirs fitted with baffle plates, in which the 
exhaust steam loses its velocity, the oil and moisture separating 
out, mainly by gravitation, as in the Baker separator (Fig. 139). 

The exhaust steam enters the separator through a branch (1) 
and is immediately caught and deflected to the lower part of the 


separator body by the baffle (2). This baffle, besides deflecting 
the steam, also tends to retain such globules of oil and water as 
adhere to it owing to the impinging of the steam against its surf ace. 
These globules eventually collect and roll down the baffle (2), find- 
ing their way into the well in the separator bottom. A large free 
passage is provided under the baffle (2), which allows of a 
decrease in the speed of the steam so that, by the time the steam 
is passing through the cleansing angles (3), it is well expanded, 
and the temperature lowered, allowing an appreciable condensa- 
tion to take place. This will be deposited on the angle bafflers 
in a chamber (4), whence the globules trickle down on to the sur- 
face of the water in the separator well (5), which is maintained 
at a constant level determined by the position of the oil pipe 
(6), through which the caught oil is discharged by gravitation if 
the steam engine is noncondensing. If it is a condensing engine, 
the oil must be pumped out by a small pump which should 
always be placed at least 24 in. below the bottom of the separator. 

It has been found sometimes that, when very high vacuums are 
carried in steam-condensing plants, the exhaust steam is not 
freed from the oil and water contained. The cause of the trouble 
needs little seeking, as the air which is always contained in the 
condensing plant, and which constantly leaks into the system, 
expands rapidly with the higher vacuums. Accordingly, the 
velocity of the vapor containing the air is so great through the oil 
extractor that any oil or water present will be swept out from the 
separator. Where means are provided so that oil and water once 
taken out of the steam cannot again enter the flow, this effect of 
high vacuums is greatly minimized. This point has been kept in 
mind in other separators which are more compact and operate 
on the principle of splitting up the steam in many little steam 
flows, frequently changing their direction and trapping the oil by 
baffle plates which are so designed that, once the oil has been 
removed from the steam, it cannot be picked up again by the 
steam but gravitates to a reservoir in the bottom, whence it can 
be removed at intervals. 

The -Princep oil separator (Fig. 140) is a good example, illus- 
trating these principles. The steam is allowed to expand and 
reduce its velocity so as to allow the solid and liquid particles to 
free themselves from the steam. A series of plates is suspended 
from the top. These plates are provided with a number of holes. 



In each hole is inserted a ferrule projecting from % to % in. 
on each side of the plate. Through these ferrules are passed 
plates twisted to a pitch equal to the distance between each 
plate. The steam on entering the separator and passing through 

FIG. 140. Princep's oil separator. 

the holes strikes the twisted blade, which, being at an angle of 
45 deg., deflects the oil and water on to the face of the plate. The 
continual action of the steam forces the deposit on the plate into 
the bottom chamber. It is impossible for any oil collected on the 
first plate to go forward to the second, because it is prevented 
from doing so by the ferrule which surrounds the hole. If the 


first deflecting action has not abstracted all the oil and sediment 
from the steam, the next or third deflection in the next chamber 
will nearly always do so; but for safety's sake a few more plates 
are fitted. 

The makers guarantee that there shall be not more than % gr. 
of oil per gallon of water (condensed steam). 

Feed Water. Chemical Treatment. Oil in emulsion can be 
removed from the feed water by adding certain chemicals 
(alumina-soda process) which produce a flocculent precipitate. 
The large precipitated particles take hold of and absorb the 
minute particles of emulsified oil, so that subsequent filtration 
easily clarifies the water. 

Feed-water softening and purifying plants are frequently in use 
where the greater portion of the feed water is taken from the town 
main or other source of fresh supply, yet in a great many large 
installations where surface condensing is resorted to, and where, 
therefore, only a small percentage of feed make-up is required, 
feed-water purifying plants may be installed with the main object 
of entirely freeing the water from cylinder oil. It becomes neces- 
sary in such cases to add a certain amount of water containing 
lime, so that, by virtue of the chemical processes, the oil may be 
thoroughly eliminated. 

Electrical Treatment. Another method is the electrical treat- 
ment in which the milky feed water containing the emulsified oil 
is passed through a tank containing two rows of iron plates; an 
electrical current is passed through the water, from one set of 
plates to the other. The result is that the minute particles of 
emulsified oil coagulate and combine with iron oxide (rust), pro- 
duced from the plates, forming a heavy deposit which can be 
easily removed by subsequent filtration through a sand filter. 

By this method it is possible to remove practically every trace 
of oil from the feed water. It is possible to guarantee less than 
0.1 grain of oil per gallon, the consumption of electric energy 
being 1 Board of Trade Unit per 1,000 gal. of water treated. 

Feed-water Softening. If certain chemicals are added to hard 
feed water containing salts of lime, magnesium, etc., some of the 
lime and oth^r ingredients are precipitated and are tak^n out in 
the form of sludge, whereas the remainder are transformed into 
such salts in solution as will not produce scale inside the boiler. 
In small plants a frequent practice is to add the chemicals to the 


hot well or directly into the feed water on its way to the boiler 
or even into the boiler itself. In this case a great deal of sludge is 
produced which necessitates frequent " blowing down" of the 

The best method of adding the chemicals is to have an inde- 
pendent feed-water softening plant, so that the water after treat- 
ment is pumped into the boiler in a condition as purified as 
possible, the sludge precipitated in the softening plant being 
removed by filtration. 

Even if the feed water has been so treated that no scale is 
being formed in the boiler, it is obvious that, as only clean steam 
evaporates away from the boiler, the water will become more and 
more concentrated with salts in solution and inclined to cause 
priming, so that a certain amount of water should be blown out 
and replaced with fresh feed water, in order to keep the boiler 
water in good condition. 

Feed water when treated is slightly alkaline ; if it is excessively 
alkaline, the boilers prime, and the degree of alkalinity should 
therefore be kept as low as possible. 


As has been explained, oil may be introduced in the feed water 
either in the form of minute particles of oil kept in suspension 
or as minute particles of water coated with a thin film of oil (oil 
in emulsion). 

When entering the boiler, the oil in suspension will rise to the 
surface more or less rapidly; and even if hardly appreciable quan- 
tities of such oil are introduced, it will almost invariably be 
noticed on the plates in the neighborhood of the water level. 
Much of this surface oil on the boiler-water level can be disposed 
of by judicious use of the scum cocks. 

The presence of oil in emulsion is, however, much more dan- 
gerous, as the small particles of emulsified oil have only a veiy 
slight tendency to rise. They combine in the boiler water with 
the solid matter, such as carbonate of lime, carbonate of mag- 
nesium, and rust (which is always introduced with the feed water 
or comes from the boiler plates). Through this combination 
with these heavier solids, the state of affairs soon becomes this: 
that the combined particles have the same gravity as the water 


and, accordingly, rise and fall with the eddy currents set up by 
circulation. They coat the underside as well as the upper side 
of tubes and flues and cling to the hot plates. The emulsified 
particles of oil which combine with the iron rust generally become 
so heavy that they sink to the bottom. 

The greasy deposit on tubes and flues has the effect of imme- 
diately retarding the flow of heat through the plate. If the 
deposit contains a sufficient percentage of oil, the flow of heat 
may be retarded to such an extent that the plate becomes over- 
heated, and the deposit begins to decompose, the layer in contact 
with the hot plate giving off various gases which blow the outer 
part up to a spongy, leathery mass, which by reason of its porosity 
retards the flow of heat even more than the thin greasy deposit. 
The plate subsequently becomes heated to redness and, being unable 
to withstand the pressure of the steam, collapses. At the same time 
the temperature has increased to such an extent that the oil is 
burned away from the deposit, leaving behind an apparently 
harmless deposit, containing the solid particles with which the 
oil originally became combined. 

It has been found that new boilers with clean flues are more 
affected by oil than are boilers in which a certain amount of scale 
is present. Many cases have been known where new boiler 
furnaces have come down when the thickness of the coating of 
grease has probably been less than 0.001 in. A coating of oil of 
this thickness will increase the temperature of the boiler plates 
several hundred degrees Fahrenheit even with a moderate rate of 

A series of experiments was carried out by the late William 
Parker, engineer in chief to Lloyd's Registry, with a view to 
determining how far the conductivity of steel and iron plates is 
affected by oil films. His experiments proved that if an open 
steel dish were painted with three or four coats of greasy deposit 
taken from the bottom of a boiler in which a furnace collapse had 
occurred, mixed with a little cylinder oil, it was possible to burn 
the bottom of the dish before the water in it boiled. 

When a boiler has become contaminated with oil, it should be 
washed out in the usual manner, then filled with water contain- 
ing 0.5 Ib. of soda ash per boiler horsepower. The water should 
be kept boiling at atmospheric pressure for 24 hr., then drawn 
off, and a thorough washing of the boiler should follow. 



Points of Application. In order to lubricate the internal parts 
of steam cylinders and valves, cylinder oil is introduced at one 
or several of the following points: 

1. Direct to the steam chest. 

2. Direct to the valves. 

3. Direct to the cylinders. 

4. Direct to the piston rod. 

5. Feeding oil into the steam line. 

1. Direct to the Steam Chest. This is one of the earliest methods 
of application. In the case of slide valves, oil is usually intro- 
duced so that it drops directly over the valve face. In the case 
of drop valves or Corliss valves (Fig. 146A), it is usually intro- 
duced at two points halfway between valves and steam pipe 
(4). The flow of steam going to the right carries along with it 
the oil to the right-hand valve (66), and the flow of steam going 
to the left carries the oil to the left-hand valve (6a). The oil 
after passing the valves enters the cylinder and provides lubrica- 
tion for the piston (1); finally it reaches and lubricates the 
exhaust valves. 

2. Direct to the Valves. Oil is delivered at one point at the 
center of the Corliss valve or at two points, one at either end of 
the Corliss valve. It is the ends of the valve that require most 
lubrication, and feeding to the ends direct is therefore preferable 
to feeding at the center, in which case the flow of steam sweeps 
the oil right through the valve without any lubrication's reaching 
the valve ends. Piston valves are sometimes lubricated by two 
oil feeds in this manner, one feed to each end of the valve. 

3. Direct to the Cylinders. Sometimes, in the case of large 
engines, oil is introduced at the center of the cylinder or at the 
top or bottom; thus introduced, it is gradually spread by the 
piston over the cylinder walls. 

4. Direct to the Piston Rod. Oil is introduced direct to the 
piston rod externally, i.e., outside the piston-rod gland, either by 
being dropped from a lubricator on to the piston rod or by an oil 
swab resting on the rod. The oil may also be introduced, par- 
ticularly under conditions of high temperature and pressure, 
directly into the piston-rod gland itself, which gives a greater 



Points of Application. In order to lubricate the internal parts 
of steam cylinders and valves, cylinder oil is introduced at one 
or several of the following points: 

1. Direct to the steam chest. 

2. Direct to the valves. 

3. Direct to the cylinders. 

4. Direct to the piston rod. 

5. Feeding oil into the steam line. 

1. Direct to the Steam Chest. This is one of the earliest methods 
of application. In the case of slide valves, oil is usually intro- 
duced so that it drops directly over the valve face. In the case 
of drop valves or Corliss valves (Fig. 146A), it is usually intro- 
duced at two points halfway between valves and steam pipe 
(4). The flow of steam going to the right carries along with it 
the oil to the right-hand valve (66), and the flow of steam going 
to the left carries the oil to the left-hand valve (6a). The oil 
after passing the valves enters the cylinder and provides lubrica- 
tion for the piston (1); finally it reaches and lubricates the 
exhaust valves. 

2. Direct to the Valves. Oil is delivered at one point at the 
center of the Corliss valve or at two points, one at either end of 
the Corliss valve. It is the ends of the valve that require most 
lubrication, and feeding to the ends direct is therefore preferable 
to feeding at the center, in which case the flow of steam sweeps 
the oil right through the valve without any lubrication's reaching 
the valve ends. Piston valves are sometimes lubricated by two 
oil feeds in this manner, one feed to each end of the valve. 

3. Direct to the Cylinders. Sometimes, in the case of large 
engines, oil is introduced at the center of the cylinder or at the 
top or bottom; thus introduced, it is gradually spread by the 
piston over the cylinder walls. 

4. Direct to the Piston Rod. Oil is introduced direct to the 
piston rod externally, i.e., outside the piston-rod gland, either by 
being dropped from a lubricator on to the piston rod or by an oil 
swab resting on the rod. The oil may also be introduced, par- 
ticularly under conditions of high temperature and pressure, 
directly into the piston-rod gland itself, which gives a greater 


certainty of its being properly distributed, as, when it is applied 
externally, the greater portion is scraped off by the gland and 
runs to waste. 

The four points of application so far mentioned are direct; 
i.e., the oil is delivered as directly as possible to the moving 
parts requiring lubrication, and, speaking generally, the more 
directly the oil is fed the less satisfactory its distribution. 

There is this disadvantage, that as cylinder oil is very heavy in 
viscosity, it spreads only with difficulty; it is apt to overlubricate 
some parts and not reach others. For this reason a great deal 
of oil is required in order to ensure that a complete lubricating 
film is maintained everywhere. 

5. Feeding Oil into the Steam Line. This is the best method of 
application and embodies an entirely different principle, as, 
instead of lubricating the various parts direct, the steam itself 
is charged with lubricant. 

By the introduction of the oil into the main flow of steam, it is 
possible to make the steam carry the oil to all parts requiring 
lubrication ; in fact, the steam itself is made a lubricant. The oil 
is introduced preferably on the boiler side of the engine stop 
valve and, in the case of saturated steam, should be introduced at 
least 18 in. away from the stop valve. 

In the case of superheated steam, which does not carry the 
oil so well as saturated steam, it should be introduced not more 
than 18 in. before the engine stop valve. In cases where the 
superheat is very high and where the steam is carried around the 
steam cylinder before it enters the valves (usually drop valves) 
on the top of the cylinder, it is not practicable to introduce the oil 
before the engine stop valve, as it would be precipitated on the 
way; it is then introduced directly into the drop valves at a 
point where the flow of steam will break it up and distribute it in 
the steam passing through the valves every time that they open. 

Atomizing the Oil. It is, however, not sufficient to introduce 
the oil into the steam pipe or flow of steam, as it is then merely 
pushed along in the form of drops. 

The best method, ensuring perfect distribution, is the atomizing 
method, by which the oil is introduced through an atomizev (Fig. 
142) into the center of the flow of steam. The steam impinging 
with great velocity (from, say, 60 to 150 ft. per second) against 
the spoon-shaped end of the atomizer will squeeze the oil through 



the slits in the atomizer, so that it is thoroughly broken up and, 
in the form of an exceedingly fine spray, mixes with the steam. 

Various atomizers have been made for the purpose of splitting 
up the oil into minor particles; e.g., it was made to ooze out from 
the perforated end of a tube, but the small holes (see Fig. 141) 

Fia, 141. Atomizer. 

divided the oil into drops only just small enough to pass through 
these holes. Other forms allowed it to be broken up over sharp 

After many trials, the author evolved the saw-slit type of 
atomizer illustrated in Fig. 142 (not patented). Its introduc- 

Fia. 142. Thomson's atomizer. 

tion has saved many thousands of barrels of oil and many 
thousands of horsepower which previously were wasted. When 
passing the slits, which should not be more than %2 i* 1 - wide, 
the oil is well atomized, and entering the engine it lubricates 


the spindle of the engine stop valve, making this valve easy to 
operate. It lubricates the steam valves and their spindles, the 
steam throwing down a slight portion of the oil on these points. 
The oil is thoroughly distributed in the form of a uniform coating 
over the piston, piston rings, and cylinder walls. The piston 
rod receives its proper share of the oil and accordingly lubricates 
the piston-rod gland packing from the inside, which is much more 
economical and efficient than lubricating the piston rod from the 

The exhaust valves receive their share of lubrication, and the 
exhaust steam, if it is carried over to the low-pressure cylinder 
(in the case of a compound engine) or to the intermediate- and 
low-pressure cylinders (in the case of a triple-expansion engine), 
will carry over finely atomized oil, so as to assist in lubricating 
these cylinders. Speaking generally, it will be found that when 
the feed of cylinder oil is ample for the satisfactory lubrication 
of the high-pressure cylinder, sufficient oil will be carried through 
to lubricate successfully the remaining cylinders. 

If between cylinders of a compound or triple-expansion engine 
there are large receivers which may perhaps be utilized for reheat- 
ing the steam, these receivers will act as oil separators, in which 
case it frequently becomes necessary to feed oil direct to the 
intermediate- and low-pressure engines; but this should be done 
by introducing it into the steam-inlet pipes leading to these 
cylinders, in preference to feeding it direct into the valves or 

Ordinarily, the oil feeds to the intermediate- and low-pressure 
steam pipes need be only from 5 to 25 per cent of the feed into 
the high-pressure steam main. 

Figure 143 shows how the two feeds from a mechanically 
operated lubricator mounted on a compound steam engine 
introduce oil to the high-pressure steam pipe and low-pressure 
inlet pipe through atomizers carrying it into the center flow of 

Where a number of engines or pumps or a row of steam ham- 
mers, each separately lubricated, take their steam from the same 
main, admirable results may be accomplished, in the way of 
saving in oil consumption combined with better lubrication, 
through the employment of one lubricator mounted on the steam 
line a good distance away from the first unit (sometimes exceed- 



ing 20 ft.) and feeding the cylinder oil through an atomizer into 
the central flow of steam. 

The steam, as previously explained, acts as a carrying medium 
for the lubricant, and each unit gets a share of the oil in propor- 
tion to the quantity of steam passing through. Atomizing the 
oil and using the steam as the oil-spreading medium results in 
the most efficient distribution of the oil, so that not only is the 
friction reduced but also the quantity of oil required for full 
lubrication. As this method relies upon the velocity of the steam 

FIG. 143. Two oil feeds for a compound engine. 

to atomize the oil, it will be understood that only in very excep- 
tional cases, where the velocity of the steam is too low, will it fail. 
This will be the case where the engines, for some reason or other, 
are operated at considerably less than half load. 

When the oil is supplied direct to the various parts, it is very 
frequently found that the piston rod, particularly under high- 
pressure conditions, is poorly lubricated. The rod shows evi- 
dence of uneven distribution of oil; it looks scratched all over 
and has the peculiar raw-polished surface that indicates wear. 
Where in such cases the atomization method is introduced, the 
oil cups furnishing lubrication to the outside of the piston rod 
can usually be dispensed with, and, owing to the better lubrica- 
tion of the piston rod from the inside, the surface of the rod will 
soon assume a glossy oily appearance, indicating that the wear 
has ceased and that the piston rod is getting a hard, polished skin. 


When stopping for week ends or for long periods, it is good 
practice to give an extra-large quantity of oil for the last 5 min. 
that the engines are running. This will give a nice coating of oil 
to all the internal surfaces and prevent the formation of rust, 
which otherwise might occur. 

Where the atomization method is introduced, it is not unusual 
to find that some of the joints between the point of entrance of 
the oil and the valve chest begin to leak, as some of the oil may 
dissolve deposits and dirt in the joints, which will therefore need 
to be tightened or repacked to keep steamtight. 

Typical Results of Using the Atomization Method. The cyl- 
inder oil should be introduced on a length of steam pipe with as 
few bends as possible before it enters the valve chest, and there 
must be no drains that might trap it. The importance of this 
point is illustrated in Example 17. 

Example 16. Four steam hammers were supplied with steam 
from the same main, and a sight-feed lubricator was mounted a 
good distance before the first steam hammer, while a drain pipe 
was fixed between this hammer and the lubricator. As long as 
all hammers were in full swing everything went well, but, when 
only one hammer was working, the flow of steam was so small 
that some of the oil was not properly atomized but dropped to 
the bottom of the pipe and was urged along it and, reaching the 
drain, dropped down. 

After the position of the lubricator was changed to a place 
between the drain and the first steam hammer, no further trouble 
was experienced. 

Example 17. On a colliery winding engine a good grade of 
cylinder oil was used, the consumption being 1^ gal. per 24 hr. 
The oil was introduced into the steam pipe but not through an 
atomizer. The Corliss valves were grinding slightly. After 
fitting an atomizer, the grinding immediately stopped, and the 
consumption was reduced to % gal. per 24 hr. work. 

After 20 months' working under these conditions the tool- 
marks on the high-pressure cylinder were not worn away, and the 
colliery manager was satisfied that no other method of lubrication 
would have kept the cylinders in such remarkably fine order. 

Example 18. A cylinder oil of good quality had been in use for 
some time with only fairly good results on a Robey compound 
horizontal engine, the oil being introduced directly into the steam 


chest by a sight-feed lubricator. When the oil was introduced 
4 ft. away from the cylinder into the steam pipe from a mechan- 
ically operated lubricator, an inspection a few weeks later showed 
that great improvement had taken place. The internal wearing 
surfaces had a nice oily appearance, and no wear was noticeable. 
It was observed, when taking out the piston, that the threads of 
nut and piston-rod end were well lubricated, whereas before they 
used to be dry, and difficulty was experienced in getting the nut 

Example 19. Two Ruston Proctor horizontal cross-compound 
engines were lubricated with sight-feed hydrostatic lubricators, 
feeding cylinder oil into the valve chest of high-pressure cylinder. 
It was necessary to resort to " flushing " on the low-pressure 
cylinders through tallow cups which were placed on the center of 
the low-pressure cylinder barrels. After altering the feed to the 
steam pipe and employing an atomizer (in this case only 4 in. 
from the valve chest, owing to a drain in the steam line), the 
consumption of the same cylinder oil was reduced 25 per cent, 
and the " flushing" of the low-pressure cylinders was found to be 
unnecessary, as the oil, atomized, was carried over with the steam. 

Example 20. Some blowing engines on an ironworks had large 
D slide valves (42 by 48 in. outside dimensions) with 8-in. travel. 
Revolutions per minute of the engine, 40; steam pressure, 60 Ib. 
per square inch ; the steam superheated to 450F. This engine 
was using J^-gal. of a very viscous mineral cylinder oil per 24 hr. 
run, and the slide valve at times jarred very badly. A com- 
pounded cylinder oil was then introduced, but although the 
valve worked better, yet it jarred badly at times, and the defect 
could be stopped only by a copious supply of oil. 

After this an atomizer was fitted, and the working of the engine 
changed at once. The valves subsequently worked very 
smoothly, the engine giving no trouble, and the consumption of 
the same cylinder oil was reduced 30 per cent. 

Example 21. A colliery fan engine (large slide valve with 
expansion valve) used 8 gal. of cylinder oil per week through 
sight-feed hydrostatic lubricators and tallow cups. 

These appliances were replaced by a mechanical lubricator, 
the feed entering flush with the inside of the steam pipe. This 
alteration made it possible to reduce the consumption of cylinder 
oil to 4 gal. per week. A further reduction was tried, but the 


amount of oil had to be increased owing to the vibration of the 
eccentric rods, which indicated that the valves were insufficiently 

Another mechanical lubricator of an improved type was then 
fitted introducing the oil through an atomizer into the same place 
as before, the result being that the engines ran smoother than 
ever, and the oil consumption was reduced to only 1% gal. per 

Example 22. On a large steam engine driving an air com- 
pressor it was found necessary to tighten the glands two or three 
times a week, when the oil was introduced direct into the valve 
chests. After the lubricator was altered to feed into the main 
steam pipe through an atomizer, the glands required to be tightened 
only once in 3 weeks. 

Example 23. A 350-hp. fan engine in a colliery consumed 
3 gal. per day of common cylinder oil fed through three mechani- 
cally operated lubricators, having a total of eight oil feeds, 
feeding direct to the Corliss valves. In addition, it was found 
necessary to feed extra oil to the ends of two of the Corliss valves, 
in order to keep them silent. 

A change was made, feeding a good-quality compounded oil 
into the high-pressure steam pipe through an atomizer, and the 
improvement in lubrication was immediately noticed. The two 
lubricators were discontinued; the consumption was gradually 
reduced to 2 pt. per day, and it was never found necessary to feed 
extra oil to the Corliss valves. 

Example 24. A two-cylinder horizontal rolling-mill engine was 
lubricated with a common straight mineral grade of cylinder oil 
internally and for the piston-rod guides. Grease was used on the 
crankpins, eccentrics, and main bearings. By the substitution 
of a good-grade compounded cylinder oil for the internal lubrica- 
tion, introduced through atomizers; and an engine oil, specially 
suited to the work, on slides, eccentrics, crankpin, and main 
bearings a great reduction was made in the power required to 
overcome the friction in the engine. 

With previous oils in use, the engine, with all load off, took 
94.2 i.hp. Five weeks after, with the new oils in use, an^ under 
exactly similar conditions, the engine consumed only 41.4 i.hp., 
showing a reduction of 56 per cent in the power necessary to 
drive it with the rolls uncoupled. The average temperature of 


slides above room was reduced from 33 F. with the old oil in use 
to 12.5F. with the new oil, showing a reduction in rise in tem- 
perature, due to friction, of 20.5F., or 63 per cent. 

The cost of lubrication was reduced by 19 per cent with the 
better grade oils in use, the actual quantity of oil required being 
only one-third of that required with the previous oil. 

The total number of indicator cards taken during both tests 
was 160, every set of cards being taken simultaneously, as all 
pencil motions were operated electrically. 

Example 25. Striking differences caused by lubrication may 
often be noticed on long-stroke, slow-speed reciprocating pumps, 
e.g., Weir's or Woodeson's type. If a change in the cylinder oil 
is made to a better grade, or if the method of lubrication is 
improved, the change will immediately result in a greater num- 
ber of strokes per minute and a smoother and more gliding motion 
of the rods, the reason being that from 25 to 50 per cent of the 
indicated horsepower is consumed by friction. 

These examples show that a decided success has followed the 
combination of mechanical lubricators with atomizers and suita- 
ble grades of oil. The arrangement must, however, in each case 
be given due thought and consideration to ensure good results. 


The Tallow Cup. The earliest form of lubricator is the tallow 
cup, consisting of an oil reservoir with a filling plug at the top 
and a cock at the bottom for emptying the oil from the reservoir 
into the cylinder or valve chest, etc. When the tallow cup is 
filled with oil, and the charge flushed into the engine, most of it 
will immediately drop to the bottom of the cylinder and be swept 
out with the exhaust steam, within the next few strokes of the 
engine. Then the engine runs on what little oil there may be 
left and within a short time will be running with no oil at all, 
until such time as the engine attendant considers it necessary to 
repeat the operation. 

When the tallow cup is fixed on the valve chest, most of the* 
oil never reaches the cylinder. It finds its way to the lower 
regions of the valve chest, mixes with any condensate that may 
be present, and is drained out. 

The tallow cup still survives as an emergency lubricator for 
flushing purposes, when extra oil is required in places where no 


oil feed is ordinarily provided for, such as top of cylinders or 
valve chests. Tallow cups are also still used for feeding oil to 
small steam pumps and the like. 

As regards proper lubricators for feeding cylinder oil, two main 
types are in use : the hydrostatic lubricator and the mechanically 
operated lubricator. 

Hydrostatic Lubricator (Fig. 144). The lubricator is usually 
attached to the steam pipe and sometimes to the steam chest. 
Steam through pipe (1) enters the condenser (2) at the top of the 
lubricator. In this condenser the steam is cooled and condensed 
into water; when the valve (3) is open, the water is allowed to 
flow down through pipe (4) into the bottom of the oil reservoir 
(5). The incoming water displaces the oil and compels it to 
flow down through the pipe (6), through the adjusting valve 
(7) fitted for the purpose of regulating the feed; then the oil 
rises through the water in the sight-feed glass (8) and enters the 
steam pipe (10) through the delivery pipe (9). 

The gauge glass (11) shows the level of the oil inside the con- 
tainer. The drain* cock (12) is fitted for drawing off the water 
before the lubricator is refilled with oil through filling plug (13). 

The distance from the steam inlet at the top of the pipe (1) to 
the top of condenser (2) should be at least 18 in. in order to 
get sufficient height of water to force the oil through the lubri- 
cator. Sometimes when a large oil feed is demanded, the pipe (1) 
is made in the form of a coil, so as to provide increased cooling 
surface for condensation. 

When the lubricator is exposed to draft or to low temperature, 
which makes the oil sluggish, it is necessary to provide additional 
water pressure by means of longer piping above the condenser. 

The lubricator must be started every time that the engine 
starts, and it must be stopped each time that the engine stops, or 
it keeps on feeding, and oil is wasted. In draining off the con- 
densed water and in refilling the lubricator, a certain amount of 
oil is usually wasted. 

The oil feed is affected by change in viscosity of the oil. It 
will therefore vary with the engine-room temperature and also 
every time that the lubricator is filled with fresh oil. T^he oil 
passes through 'small passages, which are liable to be partly 
choked with dirt, thus reducing the oil feed. For these reasons it 
is difficult to maintain a uniform feed with a hydrostatic lubrica- 



tor, more especially where a very small feed is desired. A uni- 
form feed of oil, however, is of great importance, as otherwise the 
steam is charged with either a large amount of oil too much or 
a small amount too little. 

FIG. 144. Hydrostatic lubricator. 

In connection with hydrostatic lubricators the following poiftts 
must be kept in mind. 

When the sight-feed glass is inclined to get smeared with oil, 
this may be caused by the oil drops' being very large or the sight- 
feed glass having too small a bore. The remedy is to fit a wider 
glass or to solder a wire on to the feed nipple, so as to guide the oil 


drops centrally, or to fill the glass with salt water or glycerin. 
The heavier specific gravity of these liquids causes the oil drops 
to rise earlier; i.e., the drops are smaller and do not touch the 

Leakages of joints and packings must be avoided, as they inter- 
fere with the operation of the lubricator, which is very sensitive. 

The lubricator must be filled completely with oil, and the con- 
denser must be given time to fill up with water; otherwise steam 
will enter the oil reservoir and agitate the oil, and what is known 
as " churning " will occur in the sight-feed glass. When churning 
takes place the lubricator must be emptied, cooled, filled afresh, 
and time allowed for the condenser to fill with water. 

The oil drops vary in size, according to the size of the nozzle, 
the gravity of the oil, and the liquid in the sight-feed glass ; ordi- 
narily it will be found that 1 gal. of oil will feed in 10,000 to 
24,000 drops. 

If the oil is fed by an unreliable lubricator, or if the oil feeds do 
not introduce the oil in the best possible manner, more oil is 
required to provide lubrication, and the lubrication will not be 
so efficient as when the oil is properly fed and applied. 

True economy in the lubrication of the valves and cylinders is 
obtained by feeding the minimum quantity of the correct grade of 
oil to the working parts with such regularity as will ensure an 
unbroken oil film between the frictional surfaces. Such economy 
can never be secured by the use of a lubricator that feeds inter- 
mittently or irregularly. 

The hydrostatic lubricator, which is still largely used in the 
United States, has in other countries been practically superseded 
by the mechanically operated lubricator. 

Mechanically Operated Lubricators. Mechanically operated 
lubricators are operated from some moving part of the engine; 
they therefore start feeding as soon as the engine starts and stop 
feeding when the engine stops, and they feed the oil in direct 
proportion to the speed of the engine. 

Mechanically operated lubricators preferably have sight-feed 
arrangements for each oil feed, so that the exact quantity of oil 
passing through the various delivery pipes can be observed. 
These lubricators should be so constructed that each feed is 
independent, subject to separate adjustment and control. Also, 
the working parts should not be liable to wear, and what is 


especially important all the working parts, valves, etc., should 
be easily accessible for inspection and cleaning. 

In order to ensure that the oil pipes, from the mechanically 
operated lubricator to the various parts of the engine where oil 
is introduced, shall be always completely filled with oil, spring- 
loaded check valves should be fitted at their extreme ends. The 
pipes are thus always filled with oil, and lubrication is ensured 
instantly the engine, and therefore the lubricator, start to operate. 
These check valves should be of the combined check-and-vacuum 
valve pattern, in order to prevent the oil from being sucked out 
of the lubricator container when a vacuum is formed in the steam 
line during a standstill. If the oil is introduced into a steam 
connection where a partial vacuum exists, i.e., before the low- 
pressure cylinder of a triple-expansion engine, it is essential that 
a valve of this description be fitted. 

Care should be taken that the valve does not leak and that 
the spring is strong enough to keep the valve on its seat against 
the vacuum which tends to open it. The construction and opera- 
tion of mechanically operated lubricators are treated in greater 
detail (page 88). 


The internal moving parts, comprising valves, valve rods, 
piston, and piston rod, are exposed to the action of hot steam, 
and, with the exception of the valve rod and piston rod, none of 
the internal parts is exposed to view, so that the condition of 
lubrication cannot easily be ascertained. The internal lubrica- 
tion of steam cylinders and valves is therefore of greater impor- 
tance and more difficult than the lubrication of the external 
moving parts. 

Slide Valve. The flat surface of the slide valve rubbing against 
the valve face is difficult to lubricate, particularly in the case of 
large slide valves. In some cases, oil grooves are cut in the valve 
or in the valve face, in order to assist the oil in spreading all 
over the frictional surfaces. 

The pressure between the valve and its face is great, partic- 
ularly with " unbalanced" slide valves. Improper lubrication 
results in abrasion and cutting; excessive leakage of steam takes 
place and wipes away the lubrication film from the valve face, 
necessitating an increased consumption of oil. Excessive fric- 


tion of the slide valve frequently makes the valve groan during 
operation, and the excessive resistance in moving the valve can 
usually be noticed by a trembling of the eccentric rod. 

When the cover from the slide valve chest is removed and the 
slide valve is examined, excessive friction is always indicated by 
a dryness of the rubbing surfaces, showing wear and streaks of 
cutting where the metallic surfaces have eaten into one another. 
It is important that the cast iron in the valve and in the valve 
face should be of slightly different quality or hardness, as, if the 
quality is practically the same, they do not work well together. 

Efficient lubrication of the slide valve produces a polished, 
glossy surface on the valve face. The valve operates without 
noise; the eccentric rod works smoothly, and when opened up for 
inspection the frictional surfaces show a complete lubrication film. 

Owing to the large flat frictional surfaces of slide valves and to 
the difficulty of getting the oil thoroughly introduced between 
them, and, furthermore, owing to the great pressure between the 
valve and its face, it will now be understood why the use of 
slide valves is limited to steam pressures of, say, 125 Ib. to the 
square inch, and a maximum steam temperature of, say, 450F., 
and also why overloading always makes lubrication difficult. 
Experience has proved that when the oil is introduced into the 
steam and is thoroughly atomized, the oil gets much better dis- 
tributed and has in many cases overcome groaning and trouble 
with slide valves where the direct methods of lubrication have 
failed to produce good results. 

Corliss Valves. The Corliss valve operates under conditions 
very similar to those of the slide valve, as it has a reciprocating 
sliding motion, only it oscillates over a cylindrical surface instead 
of moving over a flat surface. 

Conditions of high temperature and high pressure, therefore, 
affect the lubrication of the Corliss valve in the same manner as 
they affect the slide valve. Bad lubrication is usually noticed 
when " feeling" the valve stems. As the admission Corliss 
valves are not positively operated during the closing period, bad 
lubrication may sometimes be indicated by the valves working 
sluggishly or even " sticking." Corliss valve engines are specially 
referred to page 385. 

As Corliss valves do not work well with steam that is super- 
heated more than, say, 100F., Corliss-valve engines are less 


frequently used nowadays, and drop-valve engines are being 

Piston Valves. There is but little pressure between the piston 
valve and its cylindrical sleeve, the pressure being mainly that 
exerted by the piston rings. Exposed to high pressure or high 
temperature, the piston valve expands uniformly, and the pres- 
sure between the piston rings and the sleeve remains the same. 
High pressure and high temperature, therefore, have little effect 
on the piston valve, nor are they themselves affected by overload, 
and consequently these valves can be operated under extreme 

The signs of good or bad lubrication are similar to those 
indicated by slide valves, but, owing to its cylindrical balanced 
construction, the piston valve is easier to lubricate. It is impor- 
tant, however, that the oil be well distributed, and, again here, 
experience has shown that this can best be done by the atomiza- 
tion method of lubrication. 

Drop Valves. The drop valve lifts from and drops on its seat ; 
consequently, no lubrication is required, except for the valve 
spindle, which usually is very long and has a very short motion 
in its guide. The clearance between the valve spindle and its 
guide is slight, so that it is important to have perfect lubrication. 

The oil on the valve spindle is stagnant and exposed for a long 
time to the high temperature. It should be of the highest quality, 
so as not to bake into a carbonaceous deposit, which might cause 
sticking of the valve. 

The oil should preferably be used sparingly and introduced by 
means of the steam, so as to be uniformly distributed. 

Piston and Piston Rings. In vertical steam engines there is no 
pressure between the cylinder and the cylinder walls, except that 
exerted by the piston rings. For this reason the lubrication of 
the pistons and piston rings in vertical engines is easier, and less 
oil is required than in horizontal engines, in which, besides the 
pressure between the piston rings and the cylinder walls, is 
frequently added the pressure of the weight of the piston sliding' 
over the bottom of the cylinder. 

In the case of large horizontal steam engines, the extra friction 
due to the weight of the piston is frequently avoided by extending 
the piston rod out through the back cover and connecting it to a 
tail-rod support. In this way, by making the piston rod suffi- 


ciently rigid, the whole or part of the weight of the piston will be 
supported by the crosshead and tail-rod guides, so that the duty 
of the piston rings becomes only that of preventing leakage of 
steam from one side of the piston to the other. 

In the case of horizontal steam engines employing highly super- 
heated steam, this arrangement will .always be found desirable 
and frequently necessary, as otherwise excessive friction and wear 
result. In tandem engines, the piston rod is for the same 
reasons usually supported between the cylinders. 

The piston rings are always softer than the cylinder, so that if 
there is any wear, the greatest wear will be on the piston rings 
and not on the cylinder walls. 

During recent years a number of piston rings have been intro- 
duced which exert pressure against the cylinder walls due to the 
action of internal springs. Where the conditions are ideal, these 
rings give good service, but they are somewhat rigid in their 
construction, so that where the movement of the piston from one 
end of the cylinder to the other is not absolutely central, experi- 
ence has proved that these spring piston rings under extreme 
conditions have caused excessive friction and heavy wear. 

It must be kept in mind that the temperature of the oil film 
is high and that excessive pressure or friction may easily destroy 
the film and produce bad results. For most conditions the old 
Ramsbottom type of split piston ring, which is very flexible, 
therefore still holds its own over a wide range of service. 

It is always an advantage to have the corners of the piston 
rings rounded off, as, if they are sharp, they act like scrapers on 
the cylinder walls and destroy the oil film. When they are 
rounded, they do not dislodge the oil film, and better lubrication 

The reason why modern piston packings of rather complicated 
constructions are not so widely used as one might expect will 
perhaps be found in the fact that in event of the center line of 
the piston and rod's not being quite coincident with the center 
line of the barrel, the flexibility of the piston packing may not 
be great enough to allow for this difference. This has led to an 
endeavor on the part of piston-packing makers and designers 
to embody in their design the quality known as "flo&ting," 
which means that the particular type of packing in use may exert 
as nearly as possible even pressure all round against the walls of 


the cylinder, quite independently and without affecting the 
piston body. This same experience has also led the makers of 
metallic packing for piston rods, etc., to allow the packing a 
little lateral movement from the rod, which prevents excessive 
friction and prevents distress of the packing and subsequent 

Example 26. The following is an interesting example illustrat- 
ing how the various types of piston packing may have a bearing 
on the lubricating conditions. Complaints were made about a 
cylinder oil in use on an 8,000-hp., three-cylinder, horizontal 
rolling-mill engine, that excessive wear showed up in the cylinder, 
the cylinder walls appearing dry, no matter how much oil was 

The engine had for several months been running on a very 
small consumption of cylinder oil and giving every satisfaction, 
the oil being fed into the three steam chests on the cylinders. 
The engine was hardly powerful enough to cope with the load. 
As the chief engineer had a suspicion that some portion of the 
steam was leaking past the pistons, the cylinders were rebored, 
and new pistons were put in fitted with a modern nonfioating 
type of piston ring, instead of Ramsbottom rings, which were 
employed previously. When the engine restarted, it was found 
to be worse than ever, and the output of the steelworks largely 
decreased. At the same time the coal consumption went up, 
and it was quite apparent that more steam was leaking past the 
pistons than under the old conditions. The reason why the 
nonfloating rings did not give satisfaction was that the axes of 
the cylinders were not coincident with the axes of the three piston 
rods and tail rods. Accordingly the piston rings at a certain 
part of the stroke were bearing hard against the cylinder barrels, 
setting up heavy friction. 

At the same time, as the rings were not moving freely enough, 
steam was leaking past the pistons in enormous quantities. 
This will also explain why the cylinder walls were dry, as the 
steam oozing past the pistons would tend to carry away the 
film of oil on the cylinder walls. After some experimenting, new 
sets of piston rings were put in. These were of a type that 
allowed sufficient come-and-go (floating) to meet the conditions 
of the engine. After this, satisfactory results were again secured 
by the use of the same oil, a very marked improvement being 


shown while dealing with the maximum load, and the coal bill 
fell to normal. 

From the designer's point of view, there are several important 
things to consider in order to reduce the amount of power con- 
sumed by friction in steam-engine cylinders. 

1. The weight of the piston itself should preferably be taken by 
means other than the wearing surfaces ; in other words, the piston 
should not be allowed to wear on the bottom of the cylinder barrel. 

2. The duty of the piston rings should be only to attain steam- 
tight working. That construction would be the best which 
accomplished this with the smallest amount of pressure between 
the piston rings and the cylinder walls. Furthermore, the con- 
struction should allow of a certain amount of come-and-go, as the 
coincidence of the center line of the piston and that of the cylinder 
barrel can never be depended upon in actual practice. 

On opening up steam cylinders for inspection, the surface 
should present a rather dull apearance, coated with a thin film of 
oil. The presence of oil can be ascertained by striking a piece 
of paper around the cylinder bore at various parts of the stroke. 
After the oil film has been wiped off, the surface underneath 
should appear bright and glossy. If any wear has taken place, 
the surface will also be bright but in quite a different way; it will 
look silvery as if raw polished with fine emery cloth, and, although 
actual scoring may not have taken place, fine streaks will always 
be found, indicating wear. This may be due to a variety of 
causes, such as unsuitable or improperly selected oil; the lubri- 
cator may be unreliable, or the method of lubrication may not be 
satisfactory; or, possibly, the oil feed has been cut too low. 

Packing Glands. The function of the packing glands used for 
piston and valve rods is to prevent steam leakage outward in 
high-pressure cylinders and air leakage inward in low-pressure 
cylinders of condensing engines. 

A perfect seal can be obtained only by the presence of a com- 
plete oil film on the rods, so that full and efficient lubrication of 
the packing glands is essential. 

Many types of piston-rod glands are in service, but they can 
*be divided into two main groups, viz.: those having soft an4 those 
having metallic packing. 

Soft-packing Glands. These are used only under saturated- 
steam conditions. The friction is always comparatively high ; and 


if the packing is screwed up hard, undue pressure is produced 
between the packing and the piston rod which results in scoring 
of the latter, after which it becomes difficult to keep the gland 

In reversible engines such as colliery winding engines and steel- 
works rolling-mill engines the reversing of the engines takes 
place by changing the position of the slide valves or piston valves 
in relation to the position of the pistons. This movement of the 
valves is done by hand in the case of small engines and by a 
special reversing engine in the case of large ones. It is obvious 
that the pull required to reverse the engine is influenced by the 
frictional resistance offered by the valves moving on their seats 
and the additional resistance of the valve rods moving in their 

Where the valve rods have been lubricated externally a 
method that is wasteful and inefficient a change to the atomiza- 
tion method of lubrication brings about a marked improvement, 
particularly noticeable in reversible engines. The valve rods will 
receive internal lubrication when inside the valve chest and, 
accordingly, will convey efficient lubrication to the packing, so 
that external lubrication can be dispensed with altogether. 

The reversing lever will be easier to operate, owing to lower 
gland friction, and this is a point greatly appreciated by the 
engine drivers ; in fact, every change in the grade of cylinder oil or 
in the method of application will be immediately noticed in the 
pull required to shift the reversing lever. 

Metallic-packing Glands. Figure 145 shows a simple design. 
Metallic packing is superior to soft packing. The gland friction 
with metallic packing is appreciably less than with soft packing, 
and there is much less danger of scoring's taking place. 

It is essential when using metallic packing that the deflection 
and movement of the piston rod take place without setting up 
any undue pressures in the packing, which should exert only a 
slight pressure against the piston rod. This is accomplished by 
ball joints and annular " floating spaces " round the packing. 

Metallic packing is always employed in the case of superheated 
steam and also in the case of high-pressure saturated steam in 
large engines. 

When the atomization method of lubrication is employed with 
saturated or moderately superheated steam, it is frequently 



unnecessary to lubricate the metallic packing direct. In the case 
of highly superheated steam, however, it is always necessary to 
have a direct feed of cylinder oil into the metallic packing. Only 
the highest grade of cylinder oil should be used for this purpose 
and should be fed uniformly and sparingly, as the excess oil 
remains stagnant in the casing, which holds the packing, and, 

FIG. 145. Metallic packing. 

being exposed to high temperature, is inclined to bake into 
carbonaceous deposits. 


Experience shows that in most cases where deposits develop in 
steam engines, the cause can be traced back to the boiler conditions. 
The deposit, if analyzed, will usually prove to be " boiler matters " 
amalgamated with a larger or smaller percentage of cylinder oil, 
decomposed oil, iron, and oxides of iron. 

Deposits Due to Dirty Feed Water. Where the feed water is 
taken from rivers, it should be taken from as clean a place as 
possible, and impurities should be prevented from entering the 
water supply. In rainy weather the rivers are swollen and 
muddy; and if dirty feed water is introduced into the boilers, 
they are apt to prime, and the impurities will be carried over 
with the steam and cause deposits. 

In India the river water contains very fine suspended matter; 
this silt is carried over with the steam when the boilers prime 
and cause deposits inside the engines. It will appear that heavily 


compounded oils have proved successful in preventing such 
deposit from caking and hardening, whereas with mineral or 
only slightly compounded oils the deposit becomes hard and very 

Example 27. A 500-hp., horizontal, tandem, compound steam 
engine, using slightly superheated steam, had been lubricated 
satisfactorily with a good grade of dark cylinder oil. After an 
economizer breakdown, trouble immediately started, and a black 
deposit developed in the cylinders. The analysis was as follows: 

Per Cent 

Water 6.0 

Oil and volatile matter 43 . 4 

Metallic iron, oxides of iron, lime, and traces of copper. 50.6 

It was found that the feed water was of very poor quality and 
contained a large quantity of impurities. However, as it passed 
the Green's economizer before it entered the boilers, the econ- 
omizer pipes had the effect of precipitating the impurities in the 
lower bends, and the feed water was pumped into the boilers 
almost clean. A sample of impurities taken from one of the lower 
bends of the economizer piping was analyzed and showed a com- 
position of oxides of iron, with a large percentage of carbonate of 
lime, silicates, and also traces of coal ash. 

When the economizer broke down, the impure feed water was 
pumped direct into the boilers, and, on the boilers' priming, the 
steam carried the impurities into the steam engine, which explains 
the trouble. 

While on the subject of superheated steam, it may not be out of 
place to mention the necessity for good control of the temperature 
of the steam. 

Example 28. In one case, trouble was experienced in a steam 
engine employing superheated steam, although the temperature 
of superheat, as indicated by the thermometer placed just in 
front of the engine stop valve, showed only 530F, 

When another thermometer was brought along it recorded a 
temperature 120 in excess of this, showing that the old thermom- 
eter, probably on account of the superheat's on occasions 
exceeding the normal, had been overheated. Such overheating 
will always produce a weakening of the bulb which means a lower- 
ing of the mercury in the stem, and the thermometer therefore reads 
too low. 


Where a steam trap is not fitted or is of insufficient capacity, 
the boiler sludge will deposit in the corners and cavities of the 
valve chest, in the clearance spaces of the cylinder, behind the 
piston rings, etc. Where the oil is introduced into the main 
steam pipe and finely atomized, the greater part of the boiler 
sludge will be swept through the engine, and the valve chambers, 
cylinders, etc., will keep cleaner than where cylinder oil is applied 

The following example shows the importance of fitting a steam 

Example 29. An oil of good quality had been used on a steam 
engine employing superheated steam and giving every satis- 
faction. Without warning, trouble began. The oil carbonized 
in the cylinders, and heavy wear of the internal surfaces was 
noticed. A sample of the black deposit was analyzed and con- 
tained the following constituents: 

Per Cent 

Lime (carried over from the boilers) Trace 

Metallic iron and oxides of iron, principally metallic iron, 

produced by wear 56 . 4 

Free oil 12.8 

Volatile matter, chiefly carbonized oil 30 . 8 

It was found that through an alteration in the pipe line some 
borings had dropped into the steam line and were urged along 
with the steam. The trouble continued for a considerable length 
of time, until the last boring had disappeared. Afterward no 
trouble was experienced, the same oil giving the satisfaction that 
it gave before. 

Deposits Due to Impurities in the Steam. The solid impurities 
in the steam are mainly two kinds: 

1. Iron oxides (rust) from the boiler, superheater tubes, or 
steam line. 

2. Boiler salts and boiler impurities carried over with the 
steam during periods of priming. 

Rusty scale may come from the superheater tubes and the 
steam pipe. The cast-iron or steel surfaces in the tubes or pipes 
' will in time be covered by a scale produced by oxidation, ^s there 
is usually a slight percentage of air mixed with the steam. Owing 
to the vibration of the steam pipes and to the expansion and 
contraction due to the temperature variations, this rust in time 


breaks loose and is carried into the engines. The iron oxide from 
the superheaters is often in the form of a very fine black dust, 
whereas the rust from the steam pipe is more coarse. The 
impurities, whatever kind they may be, when entering the steam 
engine adhere and cling to the oil film all over the internal rubbing 
surfaces. The result is the formation of a dark-colored sludge or 
paste, which accumulates in the valves, valve ports, and passages; 
the spaces between and behind the piston rings ; and on the piston 

In extreme cases the piston rings will be completely choked 
with deposits; they become inflexible in their grooves; they no 
longer perform their duty of preventing leakage of steam from 
one side of the piston to the other; and the result is excessive 
wear of the piston rings and the cylinder, also heavy loss in power 
due to the increased friction and steam leakage past the piston. 
The valves and pistons groan, and the various indications of 
excessive friction characteristic of the different kinds of valve 
motion will become apparent. 

When using saturated steam, and particularly wet-saturated 
steam, the washing effect of the wet steam has a tendency to 
remove the deposits from the high-pressure cylinder and valves, 
but they are then frequently found in the passages leading from 
the high- to the low-pressure cylinder or in the latter. 

Sometimes a liberal supply of oil or the use of a light-bodied 
compounded cylinder oil will temporarily relieve the distress of 
the engine. 

In the case of superheated steam, the deposits formed in the 
high-pressure valves, valve chambers, and cylinders, particularly 
when very heavy-viscosity dark cylinder oils are used, remain 
there and are baked into hard, carbonaceous deposits, which are 
most objectionable and cause heavy wear. A liberal oil feed will 
only accentuate this trouble, as the excess oil simply decomposes 
and forms more deposits. The use of a light-bodied compounded 
filtered cylinder oil will frequently help to loosen the deposits and 
remove them from the high-pressure valves and cylinders. 

In many cases where heavy carbonization has been experienced, 
great improvements have been brought about by introducing the 
atomization method of lubrication. It is obvious that, where oil 
is introduced direct to the various frictional surfaces, it takes 
time for it to spread; therefore more oil is required, and it is -to 
this surplus oil that the impurities particularly adhere. Where 


the cylinder oil is thoroughly atomized with the steam, it is 
spread to the best advantage over the internal surfaces; it 
presents only a thin lubricating film, and there is no surplus oil 
to which the impurities can adhere. Better atomization and 
distribution of the cylinder oil therefore not only results in greater 
economy but also means cleaner lubrication internally, i.e., less 
formation of deposits. 

Where the steam is very pure, carbonization seldom occurs 
when good-quality oils are used even if the oils are fed direct 
and not atomized. If, however, the steam is dirty, the impurities 
adhere to the oil film, and because of the high temperature, a layer 
of oil carbon will be formed by oxidation. Later, a new layer of 
impurities will cover the layer of oil carbon, and another layer of 
oil will produce more oil carbon, so that if a crust of carbonaceous 
matter is examined, it will frequently be seen to consist of alter- 
nate layers of impurities and oil carbon. 

Compounded filtered cylinder oils of good quality will produce 
practically clean lubrication, notwithstanding dirty steam; such 
oils prevent the impurities from caking together with the oil, so 
that they are swept out of the cylinder with the steam. 

Where the feed water is treated chemically, and where a surplus 
of soda reaches the boiler, and priming occurs, even a small 
quantity of soda in the steam will have a very deleterious effect on 
lubrication. The soda dries up the oil film, and a more liberal oil 
feed is required when using saturated steam, whereas in the case 
of superheated steam a greater feed of oil will usually mean more 
trouble and increased formation of carbonaceous deposits. 

When reheaters are installed between the high-pressure and 
lower stage cylinders, the oil may be carbonized in these reheaters; 
and if some of it is carried over, it will cause deposits in the inter- 
mediate- or low-pressure cylinders. 

Example 30. The following analysis of a deposit taken from 
the valve chest of a 1,000-hp. horizontal steam engine is typical 
of deposits due to priming of boilers. 

Per Cent 

Iron and iron oxides 2.3 

(This represents slight wear of the internal surfaces and 

rust carried to the engine from the steam line) 
Carbonate of soda, caustic soda, and carbonate of lime . . 44 . 2 
(This has come from the boiler) 

Oil and volatile matter, chiefly oil 49 . 2 

Wfl.tfir 4 .3 


Example 31. On a colliery where the water used for boiler 
purposes was hard, the practice was to introduce soda directly 
into the boilers. Owing to this, and also to the fact that the 
boilers were worked rather at overcapacity, priming frequently 
occurred. It was found that when the steam was very wet and 
carried water containing boiler solids in suspension and various 
soluble salts, all these solids deposited themselves in the bottom 
bends of the superheater tubes, the water evaporating. When 
priming of the boilers ceased, the steam going through the super- 
heaters carried the dry dust in the bottom bends into the steam 
engines, where the deposits had the effect of " drying up" the oil 
film, so that the piston rods appeared dry; groaning of valves and 
pistons was noticed and could be stopped only with a very copious 
supply of cylinder oil. This was introduced into the main steam 
pipe through atomizers. Owing to this, quite a large percentage 
of the deposit was swept through the engine with the exhaust 
steam and into an exhaust-steam turbine. The oil and the 
boiler solids deposited themselves on the turbine blades and 
necessitated frequent cleaning, at the same time decreasing the 
efficiency considerably. 

When a feed-water softening plant was installed, the priming 
of the boilers was entirely overcome, and the troubles ceased. 

A chemical analysis of deposits developed in steam engines will, 
as indicated in the examples, always be of service in tracing 
their cause. A very simple test which can easily be carried out is 
to take a portion of the deposit and burn it on a hot plate. The 
oil will burn away, and the residue, if consisting mainly of iron 
and rust, will indicate that rusty matters have been carried over 
to the engine or that wear is taking place; if the residue consists 
of chalky matters of a light color or of a yellowish-reddish color, 
it indicates priming of the boilers, the boiler salts being carried 
over with the steam into the engine. If the whole of the 
deposit burns away, it shows that the oil in use has produced 
oil carbon and that either it is an unsuitable quality of oil, or 
the oil is used in excess or is not distributed in the best possible* 

To avoid priming it is important that the feed-water softening 
plant shall be in good working order and that the tendency of 
the boiler to prime be overcome or minimized by keeping a proper 
water level, by keeping the water in the boiler in good condition, 



and by having sufficient boiler capacity, so that the boilers are not 


In the following the lubrication of Corliss valves will be briefly 

Figures 146A and B illustrate the high-pressure cylinder of a 
steam engine having Corliss valves. The piston (1) is shown in 

A B 

FIG. 146. Corliss valve lubrication. 

Fig. 146A as moving toward the left, the steam being exhausted 
through the exhaust valve (2) to the exhaust pipe (3) leading to 
the low-pressure cylinder (possibly through a receiver). The 
steam coming from the steam pipe (4) into the valve chest (5) 
enters the cylinder, alternately passing the admission valve 6a 
or 66. 

Figure 146-B shows a cross section of the cylinder and the valves. 
The admission valve (66) is operated through the spindly (7) by 
means of the lever (8). The valve will require lubrication on the 
entire surface in contact with the valve face. How is this best 


The first attempt made to lubricate a valve of this description 
was by feeding the oil direct into the center of the valve, as shown 
by (9) (Fig. 146 B). What happened, however, was this: The 
oil that dropped on to the center of the valve was immediately 
swept through the valve-port opening. Although the valve 
needed to be lubricated along its entire length, the oil was not 
given a chance to do so and succeeded in lubricating only a 
narrow strip of the valve and valve face just in the center. 

A slight improvement on this system is feeding the oil at the 
points (10) and (11) instead of feeding it at the center. But in 
this case, also, the steam will sweep the drops of oil through the 
valve ports and prevent the oil from spreading over the entire 
valve face. The system is therefore not by a long way satisfac- 
tory, although it is advocated by the majority of engine builders. 

Where, however, Corliss valves are very big, or where the steam 
is not very clean, or in cases of superheated steam, all sorts of 
difficulties and trouble may occur. The valves groan and wear. 
They may even stick, refusing to move, causing serious irregu- 
larities in the working of the engine. The cause of the trouble is 
bad lubrication, particularly of the two ends of the valves, the 
valve end rubbing hard against the end cover. It is quite evi- 
dent that if it is difficult for the oil to remain on the middle part 
of the valve, it will be even more difficult for it to reach the two 
ends of the valves, where it is most needed. 

Probably steam will constantly keep condensing and will reach 
the valve ends but will tend only to wash away any oil that may 
be present, except when the steam itself has been thoroughly 
lubricated and therefore practically becomes a lubricant. In 
order to get the best results, the steam must be lubricated. In the 
illustration, a double-feed mechanical lubricator (12) is mounted 
on the engine, actuated by some part of the valve mechanism, 
and discharging cylinder oil through pipe (13) leading to the check 
valve (14), the drops of oil trickling down inside the atomizer (15) 
being exposed to the central flow of steam. 

In this way every drop of oil will be divided into thousands of 
the most minute particles and will be intimately mixed with 
the steam, so that when the steam is admitted through the admis- 
sion valve (6a) or (66) it sweeps over the valve faces and seats 
and will deposit sufficient oil to lubricate thoroughly. Further- 
more, some of the oily steam will condense and carry oil to both 


ends of the valves and to the valve end rubbing against the valve 
cover. Oil pipe (16) carries oil to the low-pressure cylinder. 

Cases have been known where it was impossible to stop the 
groaning of a Corliss valve even with a feed of 120 drops per 
minute of good cylinder oil, and where the mere change of the 
oil feed from feeding "direct" on to the valve to feeding into the 
steam pipe had an almost immediate effect of silencing the valve 
and doing this on a consumption of between 1 and 2 drops 
per minute. It is the old story over again, that "a drop of oil in 
the right place is better than a gallon on the floor." 

If the steam has free access to one end of the valve, and the 
access to the other end is restricted, wobbling of exhaust valves 
may occur at each stroke of the engine. The cause for this will 
be readily understood. 

Knocking of the valve-operating motions may be due to 
improper lubrication of the valves but may also simply be pro- 
duced by a loose joint somewhere. This can easily be detected by 
flooding one bearing after another of the external motion with 
oil. When the bearing that caused the knocking is excessively 
lubricated in this way, the knock, which ordinarily is sharp, will 
be deadened, as the thicker oil film in the bearing will cushion the 
blow. Adjustment of the bearing in question should therefore 
generally overcome the trouble. 


The lubrication of colliery winding (hoisting) engines presents 
several interesting features. Many winding engines are inter- 
nally lubricated by means of hydrostatic sight-feed lubricators 
feeding the cylinder oil either into the valve chest or valves or 
into the main steam pipe. Winding engines are generally hori- 
zontal twin steam engines, the main steam pipe branching off to 
each engine; and usually a sight-feed lubricator is mounted on 
each branch pipe between the throttle valve and its respective 
engine. As winding engines work intermittently, it will be 
understood that when sight-feed lubricators are in use a good 
portion of the cylinder oil will be wasted, as they continue to deliver 
oil during the periods when the engine is standing. As the sight- 
feed lubricators are generally mounted between the throttle 
valve and the engine, it will in such cases be found difficult to 



operate the throttle valve, and the valve stem will be found subject 
to more or less wear owing to lack of lubrication. 

Hydrostatic lubricators are seldom equipped with atomizers, so 
that the drivers of winding engines generally complain of diffi- 
culty in operating the reversing lever, owing to heavy friction 
in the valves and glands. 

It will also be found that in order to minimize wear on the valve 
stems and piston rods it becomes necessary to swab the rods or to 
lubricate them through a sight-feed drop oiler, dropping cylinder 
oil on the rods outside the glands. This is, of course, very 
wasteful, as most of the oil is scraped off the glands and runs to 

A B 

FIG. 147. Lubrication of colliery hoisting engines. 

When a mechanical lubricator is used feeding the oil into the 
main steam pipe before the throttle valve, through an atomizer, 
one oil feed will do to supply all requirements for the internal 
lubrication of throttle valve, reversing engine, and two cylinders, 
if the steam pipe comes to each cylinder by an equal branch, as 
in Fig. 147A, in which (1) is the lubricator feeding cylinder oil 
into the steam pipe at (2). But two feeds are necessary if the 
steam pipe is arranged as in Fig. 1475, for the greater inertia and t 
density of the cylinder oil compared with that of the steam carries 
it past the branch pipe (4) of the near cylinder, most of the oil 
being carried to the right-hand engine. 

Ordinarily, therefore, a two-feed lubricator should be fitted, 
feeding into the branches (4) and (5), respectively, at the points 
(2) and (3). 


If it is considered necessary to lubricate the throttle valve (6) 
automatically, an extra feed can, of course, be put in to deal with 
the throttle valve at (7); but if this valve is of the equilibrium 
type, a swab with cylinder oil on the valve rod over week ends will 
suffice to keep gland and valve stem in good order. 

The advantages resulting from this manner of applying the 
right grade of cylinder oil are many. 

1. There is no waste of oil, as it is fed into the main steam pipe 
in direct proportion to the number of revolutions made by the 
engine. The lubricator stops feeding when the engine comes to 

2. As the oil is properly atomized and distributed throughout 
the body of the steam, the main stop valve and the throttle 
valve will be lubricated and therefore easier to handle, the wear 
will be overcome, and the reversing engine will need no separate 

3. Each engine will receive its portion of the oil required for 
satisfactory lubrication, and it will be found unnecessary to use 
the tallow cups which are often employed to give an extra dose of 
cylinder oil direct into the cylinders when the oil is not properly 

4. As the steam is thoroughly lubricated, the valve rods and 
piston rods when coming inside the steam chest or cylinder will 
be coated with a good film of oil and thus receive their share of 
lubrication, which, in turn, will mean better lubrication of the 
gland packing, whether metallic or soft. Accordingly, less wear 
of the piston and valve rods will be apparent, and the packing will 
have a longer life. It will generally be found unnecessary to 
apply cylinder oil externally to the rods. 

5. Owing to the better lubrication of the valve glands and 
of the valves, the reversing lever will be easier to operate; and 
this is a point greatly appreciated by drivers of winding engines. 

6. Owing to the better lubrication, which means less power 
consumed in overcoming the friction, the engine drivers find that 
they can shut off steam earlier when the cage is nearing the end 
of its journey, and they also find that they can accelerate the 

'engines and the cage more quickly or with less opening ,of the 
throttle valve. 

Much the same remarks apply to steelworks rolling-mill 
engines, which also work intermittently and usually are reversing. 



The Stumpf engine has one cylinder only; steam of high pres- 
sure and high superheat expands right down to the condenser 
vacuum, the exhaust taking place through the piston's uncovering 
the exhaust port in the center of the cylinder. 

There are thus no exhaust valves; the piston is very long, so that 
the exhaust ports are uncovered, only at the right moments. 
After the steam has been exhausted, and the piston moves back, it 
compresses the remaining steam, and the clearance space when the 
piston is at the end of its stroke is very small, the intention being 
that the compression should rise quite up to the boiler pressure. 
As the steam always exhausts through ports in the center of 
the cylinder and always enters at each end alternatively of the 

cylinder, the temperature of the cylinder 
ends will be very high, and that in the 
center very low, the steam always flow- 
ing in the same direction hence the 
name uniflow engines, as they are often 

FIG. 148. Faulty uniflow- The Stumpf engine will give the same 

engine diagram. 1. Boiler re. _v i 

pressure. 2. Atmospheric efficiency as an ordinary compound 
line. 3. Maximum compres- engine using superheated steam. 

sion pressure. Qwing ^ ^ ^^ clearance? great 

accuracy is necessary in manufacture and adjustment, and the 
valves must not leak. 

The diagram (Fig. 148) is taken from a uniflow engine suffering 
from two faults, viz. y too small clearance (at that end of the 
cylinder) and leakage through the admission valve. It will be 
seen that the piston during the compression stroke compresses 
the steam that leaks in to a point far above the boiler pressure, 
partly because of the clearance spaces' being smaller than 
intended. The effect of this high compression is that the steam 
in the compression space is heated far above the normal tempera- 
ture; it may reach as high as 700F., which has a bad effect 
on the piston rod and the metallic packing. The piston rod may 
become so hot that tlie oil fumes and carbonizes badly. 

The best method to lubricate the Stumpf engine is by a six-feed 
mechanically operated lubricator, distributing the oil feeds as 


1. One feed into the steam main before the stop valve, feeding 
through an atomizer. 

2, 3. Two feeds, one into each of the vertical steam pipes, also 
through atomizers. 

4, 5. Two feeds, one into each of the admission valves, as on 
light load the oil fed into the steam pipes will not be atomized and 
reach the cylinder in sufficient quantity. 

6. One feed into the metallic packing of the piston rod. 

As these engines run at a high speed, the oil from the crosshead 
is likely to be splashed on the piston rod and get carried into the 
packing where it carbonizes. It is, therefore, always advisable 
to use cylinder oil for lubrication of the crosshead and sometimes 
also for the guides, unless special precautions are taken for pre- 
venting the bearing oil from getting on to the piston rod (see 
Fig. 82, page 254). 


Marine steam engines are often poorly lubricated. This is 
because, in times gone by, disastrous accidents and troubles with 
boilers have occurred when the cylinder oil used for internal lubri- 
cation has been carried into the boilers. Instead of endeavor- 
ing to obtain full lubrication and yet avoid boiler troubles, 
marine engineers have gone to the other extreme and have, except 
in the case of engines employing superheated steam, confined 
themselves to swabbing the piston rods and valve rods only, 
with a liberal supply of cylinder oil through the tallow cups, 
when acute trouble made it necessary to apply this .remedy. 

By the well-known practice of " swabbing the rods" most of 
the cylinder oil is scraped off by the glands and runs to waste, 
and only very little oil gets past the packings inside the engine, 
with the result that, at best, only the lower parts of valve 
chambers and cylinders are lubricated, and only very inefficiently. 

Usually, the swab pot has an open top and is exposed to coal 
dust, dirt, and impurities, which may well give rise to trouble. 

The virtue of well-lubricated valves and pistons is not only 
that the frictional losses are reduced but also that an oil seal 
is provided on the rubbing surfaces, which prevents or mini- 
mizes leakage of steam past the valves and pistons. 

Marine engines employing superheated steam cannot operate 
without lubrication. Mechanically operated lubricators are 


provided, which feed the oil into the main steam pipe or direct 
to the valves, cylinders, and packings; and if proper care is taken 
in selecting the correct quality of oil and in extracting it from 
the exhaust steam before it reaches the boilers, complete and 
efficient lubrication can be obtained without any danger of boiler 

It would be desirable to employ this same system for marine 
engines using saturated steam. The quantity of oil required is 
very small indeed, and the best results are certainly obtained by 
feeding the minimum quantity of the correct grade of oil into the 
main steam pipe before the engine stop valve and with such 
regularity as will ensure an unbroken oil film between the fric- 
tional surfaces. 

Extraction of Oil. Exhaust-steam Oil Separator. Whereas 
exhaust-steam oil separators for a long time have been in very 
general use on steam-engine plants ashore, they have not yet 
gained the same universal favor among marine engineers. 
Exhaust-steam oil separators have, however, been designed which 
are compact and suitable for marine service. 

If the bulk of the oil from the exhaust steam is removed by 
means of an oil separator, the result is that practically no oil is 
left in the steam to settle on the condenser tubes; this is a great 
advantage, as oily deposits in the condenser greatly impair its effi- 
ciency. It also means that the oil filters will more easily take care 
of the remainder of the oil and will not need cleaning so often. 

Where the internal surfaces are well worn together, and a good 
skin produced, it is sometimes possible, without any apparent 
inconvenience or trouble (owing to the wet steam generally 
carried), to operate marine steam engines for long periods without 
internal lubrication; but the internal friction is considerably 
higher than when proper lubrication is employed, and the wear 
produced on the piston rings often makes itself apparent by pro- 
ducing sharp edges, so that the rings act. as scrapers on the 
cylinder walls, producing heavy wear all round. Furthermore, 
when no oil is used internally, the leakage of steam past the piston 
rings is often considerable. 

Example 32. A remarkable instance was reported in Power 
for July 21, 1908. Four first-class armored cruisers of the 
U. S. Navy were put out of commission in a period of less than 
10 months by burned-out boiler tubes. A thorough inspection 


of the main engines showed that only a very ordinary amount of 
oil was in the exhaust steam. Examination of the auxiliaries, 
however, disclosed the trouble, which was located in the exhaust 
from .six lOO-kw.-capacity lighting sets, which were in operation 
day and night. No lubrication was used in the cylinders, but a 
careful test showed the presence of 2.2 oz. of oil per hour in the 
exhaust from each engine. These engines were of the forced- 
lubrication enclosed type, and the oil was drawn up from the 
crank chamber and crept along the piston rods into the cylinders. 
When this trouble was overcome by lengthening the distance 
pieces between the cylinders and the crank-chamber top, and no 
oil was found any longer in the exhaust from these engines, a great 
drop in the economy was at once noticed, the steam consumption 
increasing to 36.3 Ib. of steam per kilowatt-hour, whereas under 
the old conditions the engines had passed the U. S. Navy require- 
ments of " a steam consumption not exceeding 31 Ib. per kilowatt- 
hour/ ' without lubrication of the cylinders. 

However, as has been explained, the cylinders were really 
getting lubrication, although the oil was only a light-bodied oil 
from the crank chambers. A series of tests was then made on 
one of the redesigned engines, to determine the effect upon the 
economy of varying quantities of cylinder oil. The trials showed 
that when the oil feed was cut very fine, the consumption of steam 
per kilowatt-hour increased rapidly. The lowest steam consump- 
tion with ample internal lubrication was found to be 29.7 Ib. per 
kilowatt-hour, compared with 36.3 Ib. per kilowatt-hour when 
the engines were operating without internal lubrication. The 
difference in the steam consumption is due partly to increased 
consumption of power to overcome the internal friction and 
partly to the heavy leakage of steam past the piston rings due 
to the absence of the oil film. Furthermore, when the film of oil 
is not present on the cylinder walls of steam engines, radiation of 
the heat from the steam more easily takes place, the oil film being 
a bad conductor of heat. 

These trials show very clearly that the economy of a recipro- 
*cating vertical engine is to a very great extent dependent upon 
proper lubrication of the cylinders. 

When this is the case with vertical engines, it is obvious that 
proper cylinder lubrication is still more important with horizontal 
steam engines. 




The oil consumption is dependent upon many conditions which 
will be briefly referred to in the following. 

Large engines require less cylinder oil per brake horsepower-hour 
than small ones. 

Horizontal engines obviously need more cylinder oil than vertical 
engines, but care should be taken not to underfeed the latter, 
even if they do not "complain/' as it means extra friction and 
loss of steam through leakage. 

Large engines without tail rods require more oil than when tail 
rods are fitted, which relieve the pressure between the piston 
and the cylinder. 

Steam Pressure and Temperature. The greater the steam 
pressure the higher the temperature ; but when oils are chosen to 
suit the temperature, the oil consumption cannot be said to be 
influenced by the steam pressure or the steam temperature. 

Superheated steam does not, as many appear to think, mean an 
increased oil consumption; speaking generally, it may be said 
that the consumption for engines employing superheated steam, 
other things being equal, need not be more in fact, may be 
slightly less than with dry-saturated steam. Where the steam 
is dirty, the oil must be applied in the best possible manner and as 
economically as possible. 

Saturated Steam. Wet-saturated steam means an increased 
demand for cylinder oil, quite independent of whatever kind of 
impurities may enter with the wet steam. 

The oil-consumption figures given below in grams per brake 
horsepower hour may be considered approximately correct for 
average conditions. The higher figures in each case apply to 
smaller engines or wet-steam conditions, while the lower figures 

Steam engines, horsepower 




Below 400 

0.6 to 0.15 

0.6 to 0.15 
0.4 to 0.05 

Above 400 


apply to larger engines or engines employing dry or superheated 
steam or vertical engines marine engines in particular. 


The object of internal lubrication in a steam engine is (1) to 
form a lubricating film between the rubbing surfaces and thus 
replace the metallic with fluid friction as far as possible; (2) to 
form an oil-sealing film in order to prevent leakage of steam past 
the valves, pistons, and gland packings. 

Only by feeding the correct grade of high-quality cylinder oil, 
specially selected to suit the operating conditions of the engine, 
applied in the correct manner, to the right place and in the right 
quantity, will the steam engine continue to operate at its highest 
efficiency and with the minimum cost of renewals and repairs. 

Perfect lubrication is therefore dependent chiefly on the 
methods of lubrication employed and the selection of the correct oil 
for each individual case. 

If too much oil is used, lubrication under saturated-steam con- 
ditions will not be any better than when the right quantity of 
oil is used; whereas under superheated-steam conditions, the 
excess oil is detrimental, leading to the formation of carbonaceous 

If too little oil is used, a satisfactory oil film will not be main- 
tained between the frictional surfaces, so that not only will 
heavy friction and wear occur but also excessive steam leakage. 

There are a few vertical engines employing saturated steam 
which can be operated without the use of cylinder oil and without 
groaning. Nonlubrication will, however, mean excessive friction 
and excessive leakage of steam past the moving surfaces, which 
will be worth many times the cost of good lubrication. 

If an oil too heavy in viscosity is used, it will not atomize readily, 
resulting in poor distribution and necessitating excessive con- 
sumption. Because of its heavy body, the fluid frictional losses 
will be higher than they need be; and if the steam carries over 
impurities to the engine, the use of such an oil will encourage the 
accumulation of deposits, particularly under conditions of high 
pressure and superheat. 

If an oil too light in viscosity is used, it will readily atomize and 
distribute itself, but it will not be able to withstand the pressure 
between the rubbing surfaces; metallic contact will take place, 


resulting in excessive wear; also, excessive leakage of steam will 
occur, owing to the rubbing surfaces' not being completely oil 

With the right-quality oil in use, correctly selected for the con- 
ditions and applied in right quantity, a satisfactory lubricat- 
ing film will be maintained on all the internal surfaces. This film 
will be maintained with a lower consumption of oil than with 
any other grade of oil. Therefore the cost of lubrication will be 
low, and the frictional losses, because of the fluid friction of the 
oil itself as well as the leakage of steam past the moving surfaces, 
will be reduced to the minimum. 

For conditions of high pressure and superheat, the use of the 
right-quality cylinder oil will also mean that, rightly applied and 
in the right quantity, the danger of the formation of carbonaceous 
deposits will be minimized, and the possibility of excessive wear 
much reduced. 

In the following pages will be examined the conditions influ- 
encing the selection of the correct grade of cylinder oil, viz., 
steam pressure, size and construction, superheat, wet steam, 
load, impurities, exhaust steam. 

Influence of Steam Pressure. High steam pressure means 
high temperature, so that, generally speaking, heavy- viscosity oils 
are used for high steam pressures, and low-viscosity oils for low 
steam pressures (low-pressure cylinders in particular). 

Influence of Size, Speed, and Construction. The weight of a 
piston increases very nearly as the cube of its diameter, but its 
bearing surface more as the square, so that large pistons in hori- 
zontal engines, when they are not supported by a tail rod, require 
very heavy-viscosity oils. Smaller pistons and all vertical cyl- 
inders, other things being equal, will be best served with lower 
viscosity oils. High piston speed, which is found in most modern 
engines, particularly superheated-steam engines, demands lower 
viscosity oils, so as to minimize the oil drag on the pistons. 

Influence of Superheated Steam. When steam of moderate 
superheat is used, it will enter the high-pressure cylinder in a dry 
condition; but during the expansion of the steam in the cylinder 
it will cool, and, toward the end of the stroke, condensation will 

In the case of highly superheated steam, it is of the greatest 
importance that the oil should be thoroughly atomized in the 


body of the steam. There is no condensation, therefore no 
washing effect on the cylinder walls. The oil remains a long 
time in the high-pressure cylinder, exposed to friction and heat; 
while, therefore, only a small quantity of oil is required, it should 
be of such a nature that it will withstand the heat without appre- 
ciable decomposition and resultant formation of carbon. 

Dark cylinder oils exposed to heat will form more carbon than 
filtered cylinder oils. The coloring matter, which is extracted 
during the filtration process, consists of very high-specific-gravity 
bituminous matter (hence the reason why filtered cylinder oils 
have low specific gravities), which evidently decomposes and 
forms carbon. 

It has been asserted by oil firms that dark cylinder oils are 
better lubricants than filtered cylinder oils. They are, as a rule, 
more viscous, which may perhaps excuse this fallacy of opinion, 
but a moment's reflection will make it obvious to anyone that 
the chief difference between filtered and dark cylinder oils is that 
the latter contain bituminous coloring matter, the greater 
portion of which is removed when filtered cylinder oils are manu- 
factured; in other words, that the higher viscosity of dark cylinder 
oils is due largely to sticky nonlubricating ingredients, which are 
liable to decomposition if exposed to heat and other influences. 

As regards compounding superheat cylinder oils, the author 
recommends a small percentage, say 4 to 6 per cent of acidless 
tallow oil, for most conditions of superheat, as the fixed oil 
improves lubrication appreciably. 

The oil becomes very thin owing to the high temperature, and 
the fixed oil improves the oiliness of a straight mineral oil; its 
presence is therefore nearly always desirable. No ill effects 
have ever been known to be caused by decomposition (formation 
of fatty acid) of such a small percentage of fixed oil. On the 
contrary, it will tend to prevent carbonized matter from baking 
together and forming hard crusts, in this way making the nature 
of such deposits less dangerous. 

Influence of Wet Steam. Where the steam is wet it has a 
tendency to wash away the oil film on the internal surface^. In 
compound or triple-expansion engines, even if the steam is dry 
on entering the high-pressure cylinder, the fall in pressure and 
expansion taking place produces condensation, so that the steam 
arriving at the low-pressure cylinder usually is very wet. 


It is obvious that the problem of lubricating the high-pressure 
cylinder under dry-steam conditions is different from lubricating 
the high-pressure cylinder under wet-steam conditions or from 
lubricating the low-pressure cylinders under very wet-steam 

In order to lubricate cylinders satisfactorily under wet-steam 
conditions, the cylinder oil must readily combine with the mois- 
ture and cling to the cylinder walls; i.e., it must be a compounded 
cylinder oil. It is therefore frequently desirable to use one 
grade of cylinder oil for the high-pressure cylinder and a different 
grade (lower viscosity, more heavily compounded) for the low- 
pressure cylinder in large compound or triple-expansion engines. 

Influence of Engine Load. The greater the engine load the 
greater the volume of steam passing through the steam pipe into 
the engine; and the higher its velocity the better will it be able 
to break up the cylinder oil introduced through the atomizer. 

As superheated steam does not atomize and distribute the oil 
so well as does saturated steam, engines employing superheated 
steam and likely to operate under light load conditions should 
have means for lubricating the internal parts direct in addition 
to introducing the oil where it can be atomized. Light load also 
means that the steam expands more in the high-pressure cylinder, 
so that at the end of the piston stroke the steam is much more 
moist (more condensation) than under full-load conditions. 
Wet steam calls for compounded cylinder oil, so that, speaking 
generally, light-load conditions demand compounded oils of low 

Influence of Impurities in the Steam. It has already been 
mentioned how iron oxides, boiler salts, etc., have the effect of 
combining with the oil and forming deposits. The higher the 
viscosity of the oil the more difficult will it be to avoid such 
deposits, as such oils cling tenaciously to the impurities. Low- 
viscosity oils are therefore to be preferred, where a great deal of 
impurities enter with the steam; this is particularly the case 
under conditions of superheat. 

As the presence of impurities in the steam usually means that 
priming of the boilers is responsible, in the first instance, the 
steam will be wet, so that oils heavily compounded are, as a 
rule, called for. There is one exception to this rule, viz., that 
under conditions of high superheat; where it is only the dry 


boiler salts that reach the engine, and where these dry salts con- 
tain alkali, e.g., soda, they will form a soap with the tallow oil 
present in the cylinder oil, which will aggravate the deposit 
trouble, whereas with a straight mineral oil such soap cannot 
possibly be formed. 

For saturated low-pressure steam conditions, there is no great 
difference between dark or filtered cylinder oils as regards forma- 
tion of deposit by impurities; but for superheated steam condi- 
tions, filtered cylinder oils are vastly superior, as, under the dry 
high-temperature conditions, the bituminous matter in dark oils 
combines with the impurities, decomposes, owing partly to oxida- 
tion, e.g., oxygen being taken from the iron oxides, and forms hard 
brittle carbon. 

Generally speaking, the presence of impurities under saturated- 
steam conditions therefore calls for oils of low viscosity and com- 
pounded (filtered oils not particularly needed), whereas 
impurities under superheated-steam conditions demand mineral 
oils of low viscosity and filtered (compounded oils may form soap). 

Influence of Exhaust Steam. As mentioned elsewhere, it is 
under certain conditions desirable to extract the oil from the 
exhaust steam and to eliminate as far as possible the danger 
arising from its getting back into the boiler. All compounded 
cylinder oils are difficult to separate from the exhaust steam and 
from the feed water. All straight mineral oils are fairly easy 
to extract, but the dark oils combine rather intimately with the 
water, forming semiemulsified clots of oil (which cannot be used 
again), and just a trace of the oil goes into a fine emulsion. 

Well-filtered straight mineral oils separate easily from the 
feed water, and the oil can be recovered and used on less impor- 
tant work; the feed water will be practically free from emulsified 

It will, however, be found that more oil is required when 
using a straight mineral cylinder oil than when using a com- 
pounded cylinder oil, so that the best results will often be pro- 
duced by using a slightly compounded filtered oil, as such an oil 
will give more efficient and more economical lubrication. The 
oil should be fed as economically as possible, so that there will 
be only a small quantity present in the exhaust steam. Dia- 
grams 1 to 6 may prove of interest in connection with the influence 
of exhaust steam on the selection of oil. 








When the exhaust steam is discharged into the atmosphere, the cylinder 
oil ma}' be chosen entirely with a view to suiting the engine requirements. 

When a contact feed-water heater is fitted, straight mineral, dark, or 
filtered steam-cylinder oils must be used. 









Straight mineral, dark, or filtered cylinder oils must be used, or filtered 
oils, slightly compounded, used very sparingly. 

With some vertical engines, aquadag has been used successfully, and the 
condensed steam from the heating system returned to the boilers. 












Straight mineral, dark, or filtered oils must be used, or filtered oils, slightly 
compounded, used very sparingly. 




'farety Fiffect) 





Straight mineral, dark, or filtered oils must be used, but when an exhaust- 
steam oil separator is fitted, filtered oils lightly compounded are recom- 
mended and will give efficient lubrication; they can and must be used very 








The oil may be chosen entirely with a view to suiting the engine require 
inents, as every trace is eliminated from the feed water in the deoiling plant 



A MO USED AGA/M, Q/? {/S> F0# 


When an exhaust-steam oil separator is fitted, the oil may be chose 
entirely with a view to suiting the engine requirements; when no oil separate 
is fitted, and when the hot-condenser water is used and comes in contac 
with textile fabrics, heavily compounded oils must not be used. 



The oil should be tested for a period of at least 3 months in 
case the first few days' working has been satisfactory. It takes 
time for a good cylinder oil to produce a good working skin on 
the internal wearing surfaces; in fact, it takes much longer than 
for an unsuitable cylinder oil to destroy the good surface produced 
by a suitable oil. After a few days, the consumption of oil should 
be gradually decreased, and the minimum feed determined by 
which smooth and satisfactory running can be accomplished. At 
the end of 3 months' working on the reduced feed the cylin- 
ders should be opened up for inspection and should present a sur- 
face of rather dull appearance, coated with a film of oil. 

The same remarks will apply to the appearance of valve rods, 
piston rods, valves, and valve faces. Whenever a change of cyl- 
inder oil is made, irregularities may be experienced during the 
earlier period of its working, owing to the new oil's altering the 
wearing surfaces. Where unsuitable oils have been in use, and 
various deposits have accumulated behind the piston rings and in 
the glands, cylinder oil of a good grade will clean the surfaces. 
In such cases dirt may be carried to the piston rod, and the new 
oil generally gets the blame. 


Before giving specific recommendations for different types of 
steam engines, it may be well to examine briefly the physical and 
chemical tests most often referred to when judging the merits of 
cylinder oils. 

These are specific gravity, viscosity, flash point, percentage 
and nature of compound, color, cold test, and loss by evaporation. 

Specific Gravity. The lower the specific gravity for oils of 
similar viscosity the purer the oil. A highly filtered cylinder 
oil will be lower in specific gravity than one less purified. It must 
be kept in mind that these statements are true only because prac- 
tically all steam-cylinder oils are produced from paraffin-base 
crudes, which are rather similar in nature. 

Viscosity. The viscosity taken at 212F. is always useful. It 
has been often referred to in the preceding pages in connection 
with "influencing conditions." 


The admixture of tallow oil reduces the viscosity but increases 
the lubricating power of the oil its oiliness. Filtered cylinder 
oils have lower viscosity than dark cylinder oils but greater fric- 
tion-reducing powers. In comparing viscosities of different oils, 
one must therefore keep in mind whether they are compounded or 
more or less filtered. 

Flash Point. Although it is true that good cylinder oils for 
use with superheated steam do possess a fairly high flash point, 
yet it is by no means certain that a cylinder oil having a high 
flash point is suitable for work with superheated steam. The 
flash point is determined in the laboratory under atmospheric 
conditions. If the cylinder oil were to be tested under the high 
pressure carried in the steam pipe, the flash point would undoubt- 
edly be shown to be considerably higher, just as the boiling point 
of water, which at atmospheric pressure is 212F., increases with 
any pressure above that of the atmosphere; (e.g., at 150 Ib. per 
square inch the boiling point of water is 366F,). This will 
explain why it is frequently possible to use a cylinder oil suc- 
cessfully for lubrication under superheated steam condi- 
tions where the temperature of the steam is even a good deal 
higher than the flash point of the oil, measured under atmospheric 

Besides, there is practically no air present in the steam, and 
therefore no danger of the oil's flashing anywhere. The tempera- 
ture of the piston or valve rods, which are the only hot frictional 
parts passing out into the atmosphere, is always considerably 
lower than the maximum steam temperature, so that the flash 
point of the oil is never reached; and even if it were reached, noth- 
ing much would happen, there being no chance of an explosive 
mixture's being formed of oil vapor and air, such as may be the 
case in air compressors. 

Compounded Oils. For most conditions, experience has 
proved that cylinder oils compounded with the proper kind and 
amount of fixed oil are more suitable than those which are straight 
mineral. It is more particularly where steam engines are work- 
ing with wet steam that the advantage of using compounded oils 
becomes apparent. Great care must be exercised in selecting the 
proper kind of fixed oil, as unsuitable fixed oils under the action of 
steam at high pressure and temperature decompose and develop 
acids and gummy residues which corrode the internal wearing 


surfaces and produce sticky, pasty deposits which unduly increase 
friction. Compounding mineral cylinder oils with the right 
proportion (from 4 to 15 per cent) and quality of fixed oil, pref- 
erably acidless tallow oil, usually adds to its lubricating value, 
and better results will be secured than if the cylinder oil is used 
without the admixture of fixed oil. 

Color. The more highly filtered a cylinder oil is the lighter will 
it be in color, so that light color (low Lovibond-color number) 
usually signifies a high degree of purity. 

Cold Test. It is desirable that a cylinder oil keep fairly fluid 
at ordinary engine-room temperatures, especially when used 
through hydrostatic lubricators, with which difficulty is always 
experienced in feeding viscous cylinder oils at a regular rate of feed. 
When good mechanically operated lubricators are employed, the 
cold test of the cylinder oil is of less importance. 

Loss by Evaporation. Such laboratory tests as determine the 
percentage of evaporation when heating a sample of cylinder oil 
to a certain temperature for a certain time are of very little value 
in determining lasting properties of a cylinder oil, as these tests 
are carried out under atmospheric pressure and under conditions 
greatly different from those met with in actual work. 

It will be understood from the foregoing that the author con- 
siders the following tests of great importance: 

Specific gravity and color as indicating degree of purity. 

Viscosity to suit conditions of temperature and pressure. 

Percentage and nature of compound to suit wet-steam condition 
and increase oiliness. 

Cold test (equivalent to viscosity at low temperatures) to 
ensure proper feeding of the oil through the lubricators. 


Acidless tallow oil and not tallow is generally used for com- 
pounding cylinder oils, because tallow is often acid or rancid and 
therefore inferior to acidless tallow oil. Tallow and black lead 
used to be a favorite cylinder lubricant at sea, when steam pres- 
sures were low; but with the advent of higher steam pressures 
such mixtures have almost disappeared. Yet it is not infrequent 
to find engine drivers of both stationary and locomotive engines 
in the habit of using tallow indiscriminately, particularly under 


wet-steam conditions; it keeps the engine quiet and makes the 
cylinder oil last longer. The acidity produced by decomposition 
of the tallow (into fatty acid and glycerin) will, however, in time 
act most destructively on all cast-iron surfaces. A symptom 
often exhibited is that the acid "perf orates " the skin on the 
piston rods; the rod then becomes pitted and wears badly. It 
also causes, inside the valve chests and cylinders, deposits com- 
posed chiefly of iron soaps and may soon cause sufficient corrosion 
and pitting to ruin the surfaces after a comparatively short life. 

In locomotives a portion of the deposits reaches the smoke-box 
exhaust nozzle and cakes, owing to the great heat, closing the 
nozzle and causing a labored exhaust until cleared away. 

Cast iron long exposed to the action of fatty acids from tallow 
becomes so crumbly that it can be cut with a knife like cheese. 
The metal is porous and filled with iron soaps, etc., which explains 
why it is so exceedingly difficult to introduce an oil, largely min- 
eral in character, where tallow (or, for that matter, any fixed oil, 
such as rape oil Colza occasionally favored by engine drivers 
for troublesome engines) or cylinder oils containing a large 
percentage, say 20 to 25 per cent or more, of tallow, tallow oil, 
or other fixed oil has been in use for a long period. 

The only way is to introduce the oil very gradually mixed with 
the old lubricant over a period of at least 3 months, gradually 
increasing the percentage so as to give the acid products of 
decomposition time to loosen, dissolve, and get cleared away 
through the exhaust. If the oil is introduced too quickly, 
it will dissolve the deposits too rapidly, with the result that 
excessive scoring and wear inevitably take place, and a return to 
the old lubricant becomes necessary if the engine is not to suffer 
more serious damage. 

In America, semisolid greases, containing more or less tallow, 
are not infrequently used as cylinder lubricants. They are more 
difficult to apply economically than a proper grade of cylinder 
oil and cannot possibly give better lubrication, as they either 
contain a percentage of nonlubricating material or, if they are 
rich in tallow and such like, give rise to troubles with corrosion 
of surfaces or with the feed water (too much compound). Weight 
for weight cylinder oil of the correct grade is always preferable, 
and besides it will be found that if the price per pound is com- 
pared with that of semisolid grease, the latter is always dearer. 



The use of such lubricants cannot therefore be recommended from 
any point of view, except perhaps that of the manufacturer. 


The lubrication chart shown on page 409 gives specific cylinder- 
oil recommendations for all types of steam engines. Before 
describing how it is to be used, it will be necessary to describe 
what the various grades of cylinder oil represent. 

Cylinder oils of four viscosity ranges Nos. 1, 2, 3, and 4 have 
been found adequate for the lubrication requirements of all 
types and sizes of steam engines. These viscosity ranges are 
shown in the following table, also the approximate specific 




Open flash 

Cold test, 




point, F. 


Grade of cylinder oil 























No 1 filtered 

11 to 13 

14 to 18 



40 to 50 

No. 2 dark, No. 2 filtered 







50 to 60 

40 to 50 

No. 3 dark, No. 3 filtered 







50 to 60 

40 to 50 

No 4 dark 





50 to 60 

* See table, p. 57. 

gravities, flash points, and cold tests, corresponding to these 
viscosities, for both filtered and dark oils. 

There is no demand for dark oils of the viscosity of No. 1 
grade, and it is not possible commercially to manufacture filtered 
oils of the No. 4 grade viscosity, nor do the actual requirements 
call for such oils. Filtered oils of the No. 3 grade viscosity are 
superior to dark oils of the No. 4 grade viscosity as regards oili- 
ness (which is more important than viscosity), but they are 
more expensive to manufacture. The various oils may also be 
straight mineral or more or less heavily compounded with acidless 
tallow oil. 

A few dark cylinder oils are marketed having higher viscosities 
and flash points than the No. 4 grade. Such oils are unneces- 


sarily viscous, waste power, and easily carbonize and form 

In the table below are indicated 12 grades of cylinder oils, 
6 filtered and 6 dark, representing the author's recommendations 
based on practical experience with such oils on a vast number of 
steam engines. 



Cylinder Oil Number 

No. 1 filtered, heavily compounded (10 per cent) 1 F.H.C. 

No. 1 filtered, lightly compounded (4 per cent) 1 F.L.d 

No. 2 filtered, medium compounded (6 per cent) 2 F.M.C. 

No. 3 filtered, medium compounded (6 per cent) 3 F.M.C. 

No. 2 dark, medium compounded (6 per cent) 2 D.M.C. 

No. 3 dark, medium compounded (6 per cent) 3 D.M.C. 

No. 3 dark, heavily compounded (10 per cent) 3 D.H.C. 

No. 4 dark, medium compounded (6 per cent) 4 D.M.C. 

No. 2 filtered, straight mineral 2 F.S.M. 

No. 2 dark, straight mineral 2 D.S.M. 

No. 3 filtered, straight mineral 3 F.S.M. 

No. 3 dark, straight mineral 3 D.S.M. 

The 12 grades in the foregoing table will be found in the first 
column of the lubrication chart on page 409. The other verti- 
cal columns refer to the conditions influencing the choice of 
cylinder oil. The black squares in each column indicate the 
condition for which the cylinder oil (shown at the left extreme of 
the same horizontal line) is not suitable. 

In order to find an oil suitable for a certain set of conditions, 
take a piece of paper and place it with its upper edge along the top 
line ; make a pencil mark on the edge of the paper corresponding 
to each set of conditions and opposite the condition found in the 
steam engine in question. It is important that a mark be made 
corresponding to all seven groups of conditions in order that the 
recommendation made by the table may be correct. Having 
marked the paper at seven places, move it down to the first 
horizontal line; if none of the seven marks clashes with (corre- 
sponds with) any of the black squares on this line, Cylinder Oil 
No. 1 F.H.C. (No. 1 filtered, heavily compounded) is the correct 
grade of oil to use. If one or more of the black squares clashes 


For Steam Cylinders and Valves 


Also Recommended for Large Low Pressure Cylinders with Wet Steam 

NOTE 1. For light-load conditions choose an oil slightly lower in viscosity and/or more 
heavily compounded than the one indicated by the chart. 

NOTE 2. With impure steam (boiler's priming, etc.) a filtered oil should preferably be 
used, and with saturated steam preferably compounded. 

NOTE 3. When the chart recommends more than one grade, the one lowest in viscosity 
should preferably be chosen; when a dark as well as a filtered oil is recommended, as will 
often be the case, the former, unless there are special conditions (Note 2), may be preferred, 
as it is (or ought to be) lower in price. 

NOTE 4. A straight mineral oil can always be used in place of the compounded oil recom- 
mended by the chart, but it means an increased oil consumption as compared with a medium- 
compounded oil of 50 to 100 per cent; the use of a straight mineral oil in place of a slightly 
compounded oil or the latter in place of a heavily compounded oil means an increase in oil 
consumption of 30 to 50 per cent. 

NOTE 5, From 10 to 15 per cent of compound may be required in case of (a) very wet 
steam in large engines, low-pressure cylinders in particular; (6) heavily loaded Corliss 
valves or unbalanced slide valves; (c) very dirty steam, particularly saturated steam. 

NOTE 6. No. 2 F.S.M. and 3 F.S.M. will separate more easily from the exhaust steam and 
feed water than No. 2 D.S.M., and 3 D.S.M. and will give cleaner and better lubrication, 
particularly under conditions of superheated steam and/or impure steam. 


with the pencil marks, move the paper down to the next hori- 
zontal line. If there are still obstacles in the way (black squares) 
move to the third line and so on until a line is found where there 
are no obstacles opposite the pencil marks. The correct oil 
will then be shown in column 1 of that particular horizontal line. 
Do not go from line 1 to line 5, because the first four lines all 
refer to No. 1 F.H.C. ; they represent different sets of conditions 
and no lines must be missed. 



From a lubrication point of view there are two main groups of 
locomotives, viz., railway locomotives, employed in more or less 
regular service on railways; and works locomotives, such as 
are employed in steelworks, mines, quarries, and shunting 

Works Locomotives. It is often painful to see the crude way in 
which lubrication is provided in most works locomotives. Many 
small locomotives are only fitted with tallow cups, and at best 
some kind of hydrostatic lubricator as a rule, the cheapest 
possible is installed. 

With tallow cups, lubrication is always poor, whether the oil 
allowance is great or small. With hydrostatic lubricators there is 
always waste of oil, as they keep on feeding, quite independent 
of the actual requirements. The drivers are not so careful as 
railway-engine drivers and do not, as a rule, trouble to shut off 
the lubricator every time that the locomotive stops for a little 
while. Mechanically operated lubricators, operated from one of 
the valve spindles, similar to stationary-engine practice, will save 
a great deal of oil on all such locomotives and provide more 
uniform lubrication than will hydrostatic lubricators. 

It is necessary to fix the mechanical lubricator with heavy 
brackets to the engine frame and to take every precaution that 
vibrations from the engines are felt by the lubricator as little as 
possible. The oil should preferably be introduced by means of 
an atomizer (see page 361) into the steam pipe in the smoke box, 
before it branches off to each cylinder. 

When the oil is thoroughly atomized, the steam lubricates 
valves, cylinders, and piston rods, so that there is no need for 


extra lubrication of the rods. But where hydrostatic lubricators 
or tallow cups are employed, it is necessary to have a swab or 
mop for the rod glands. Such swabs are made of worsted or 
cotton (lamp wicks), plaited and formed into a ring, placed round 
the rod and held in position by the gland nuts; they are preferably 
enclosed in a box to protect them from dust and grit. 

Railway Locomotives. -Coming now to the other and more 
important group of locomotives those employed in regular rail- 
way service, whether passenger or freight we find that there is 
one condition that vitally affects the lubrication question, viz. 9 
that when a train passes a down-gradient portion of the line, the 
steam is practically shut off; i.e., the engine is what is termed 
" drifting" with a closed throttle. If the oil under these condi- 
tions were introduced into the steam pipe, there would be no 
steam to carry it into the valves and cylinders; and if the down 
gradient were a long one, the rubbing surfaces would soon be 
devoid of lubrication. 

During periods of drifting, another complication occurs; the 
valves and pistons act like pumps and may create a vacuum 
ranging from 3 to 9 Ib. on the exhaust side which sucks ashes 
and soot into the cylinders from the smoke box. These impuri- 
ties adhere to the cylinder oil and may form very objectionable 
crusty deposits in the valves, passages, and cylinders. To over- 
come this difficulty, good practice requires either that the driver 
shall very slightly open the regulator when the engine is drifting 
or that a by-pass valve (snifting valve, antivacuum valve) be 
provided, which automatically admits sufficient steam to the cylin- 
ders so as to kill the vacuum and prevent the entrance of soot 
and ashes. Some snifting valves are designed to admit air 
instead of steam or air and steam. This practice is permissible 
for saturated steam, but with superheated steam the internal 
temperatures are so high that the air immediately oxidizes the 
oil and causes the formation of sticky, carbonaceous deposits. 

It will now be realized that the condition of " drifting" neces- 
sitates the oil's being introduced straight into the valves and 
cylinders. With saturated steam an oil feed to the cylinder is 
seldom required, but with superheated steam the cylinder feed 
cannot be dispensed with. f 

Speaking generally, 75 per cent of the oil is preferably intro- 
duced into the valve chest, and 25 per cent into the cylinders. 



As to the method of introducing the oil, there can be no question 
of the superiority of the atomization system over all others, 
and for superheated steam conditions in particular, as will be 
explained presently. 


Both hydrostatic displacement lubricators and mechanically 
operated lubricators are employed and there have been great 
controversies of opinion as to their respective merits. 

Hydrostatic Lubricators. These lubricators are fitted in the 
cab, as shown in Fig. 149. Steam is admitted to the lubricator, 

FIG. 149. Hydrostatic locomotive lubricator. 

condensing in the upper part it; by gravity displacement the 
oil is forced up through sight feeds, and through long feed pipes 
it finally reaches the valve chests and cylinders. The best hydro- 
static lubricators admit saturated steam to the feed pipes. The 
steam keeps the pipes hot and more or less emulsifies the oil, so 
that it is readily atomized in passing through the choke plug C, 
always fitted before the oil enters the engine. Figure 150 shows 
in detail such a choke plug; a valve (1) is kept constantly vibrating 
on its seat by the motion of the engine; the mixture of oil and 
saturated steam passes through fine channels and cross channels 
in the valve or between the valve and its seat; the churning 
action thoroughly atomizes the oil; in fact, what is produced is 
really oily steam " Scotch fog" which spreads quickly over 
the internal surfaces and forms the best means by which the oil 
can be distributed. 

If the choke plugs were absent, the difference between the 
boiler pressure and the pressure in the valve chest or cylinder 



would cause waste of steam through the oil feed pipes, particu- 
larly when drifting. The choke plugs are therefore required for the 
dual purpose of checking the steam flow and atomizing the oil. 

When applied to locomotives employing saturated steam, two 
feeds, one for each valve chest, will suffice for 
most high-pressure engines ; but the cylinders in 
large engines will occasionally be better lubri- 
cated if they are lubricated direct, so that a four- 
feed lubricator is required. An extra feed may 
be added for feeding the air-pump cylinder. 
This oil feed must not have steam admission ; the 
oil drops through a sight feed and gravitates to 
the air cylinder. 

For superheated-steam conditions, hydrostatic 
lubricators are used almost exclusively in the 
United States and Canada. Some British rail- 
ways are also using them and getting good 

Although the lubricators first fitted had a great 
number of feeds, it seems now to be an estab- 
lished fact that for all two-cylinder engines one 
feed into each valve chest (into the middle with 
inside steam admission or a divided feed into FIG. 150. Choke 
both ends with outside steam admission), one plug c ^ Flg * 149 ^- 
feed into each cylinder, one feed divided to the tail rods, and one 
feed for the air pump, making six feeds in all, will provide proper 
oil distribution. For four-cylinder engines more feeds are 
required, and it is advisable to fit two lubricators, one for either 

In the United States the oil feeds on each side are often 
divided to serve both valve and cylinder, but in view of the uncer- 
tainty as to which path the oil will choose, it seems better practice 
to feed the valve and cylinder by separate feeds. If feeds are to 
be divided, it would be better to divide one for both valves or 
for both cylinders, as with this arrangement one may with better 
reason expect a fair distribution of the oil. 

The division of feeds must, of course, be done after the oil has 
passed the choke plugs. As to British practice, at least one rail- 
way has divided the cylinder feed without any apparent ill effects, 
but the feeds to the valve chests have not, to the author's knowl- 


edge, been divided. As the greatest amount of oil has to be fed 
to the valves, this practice appears to be sound and preferable to 
the American one of dividing the feeds, which certainly introduces 
an element of uncertainty. 

Mechanically Operated Lubricators. Mechanical lubricators 
have a container from which the oil pumps draw the oil; the con- 
tainer, therefore, is not under pressure and can easily and quickly 
be refilled with oil. Filling a hydrostatic lubricator with oil is 
more complicated, as the water first must be emptied out, and 
there are several valves to look after every time to ensure correct 
working of the lubricator when starting up again. Mechanical 
lubricators start feeding as soon as the engine starts and stop 
feeding with the engine, so that no oil is wasted while the engine is 
standing. Hydrostatic lubricators must have their oil feeds 
started about 10 min. before the running, and they keep on 
feeding while the engine is standing or running slowly. 

Mechanical lubricators feed the oil according to the speed of 
the engine, whereas a hydrostatic lubricator will feed approxi- 
mately the same amount of oil whether the engine goes fast or 
slow, whether on an uphill or a downhill gradient. When super- 
heated steam was first introduced on the Continent, mechanical 
lubricators were thought necessary; the principle of atomization 
was not understood or appreciated, and as a result the great 
majority of locomotives in Europe, South Africa, India, and the 
East generally are fitted with mechanical lubricators without any 
attempt's being made to atomize the oil. Numerous troubles 
with excessive carbonization, heavy wear, and friction are 
recorded, too numerous to be disregarded. 

What happens is that the oil is injected unatomized into the 
valves and cylinders; it is very viscous and spreads only with 
difficulty; it is exposed to high temperature, to the oxidizing 
effect of hot smoke-box gases and boiler impurities, and to con- 
tamination from soot and ashes. As a result, particularly if the 
oil consumption is liberal, very tenacious sticky or hard carbon- 
aceous deposits are formed. The rubbing surfaces become poorly 
lubricated, and heavy friction and wear take place. Frequent 
cleaning of valves and cylinders and keeping the oil consumption 
as low as possible will assist in preventing trouble, but even 
with the best possible attention to these points it is difficult to 
ensure perfect lubrication. 



Of course, if suitable antivacuum valves are fitted, if the boiler 
water is of good quality, and priming only slight or absent, it is 
possible to get good results with mechanical lubricators. But 
results in practice generally fall short of perfection, and it is under 
more or less unfavorable conditions that feeding the oil unat- 
omized is almost sure to give trouble. The fault is not with the 
mechanical lubricators themselves. Stationary practice has long 
since proved that they are superior to and more economical than 


Fio. 151. Thomson's atomizer arrangement. 

hydrostatic lubricators; the cause of most carbonization troubles 
is simply that the oil is not atomized. 

The good results obtained with hydrostatic lubricators under 
superheated steam conditions have proved that if the oil is intro- 
duced as oil fog, saturated steam being the carrying medium, 
it has the effect of keeping the rubbing surfaces free from deposit. 
Whatever impurities may be drawn into the engine during periods 
of drifting are prevented from caking and are expelled through 
the exhaust when steam is again admitted. 

Experience has proved that perfect atomization is imperative, 
if carbon deposits are to be avoided with superheated steam. 

The author believes that he was the first to suggest the com- 
bination of mechanical lubricators with atomizing boxes (in a 
paper read before the Institution of Locomotive Engineers, 
London, on Mar. 25, 1915). Figure 151 shows the author's 




design, which has proved efficient in overcoming carbonization 
troubles. The feed pipes (1) from the mechanical lubricator L 

discharge oil through check 
valves (2) into the atomizer 
box (3), shown in detail in Fig. 
152. Saturated steam is sup- 
plied through an auxiliary pipe 
(4) and causes the oil to be pre- 
liminarily atomized through 
the saw slits of the atomizer 
(5); the mixture of oil and 
saturated steam is finally 
atomized in passing through 
the choke plug (6). 

It will thus be seen that the 
steam has an unobstructed flow 
through the atomizer box and 
that each feed gets its fair share 
of the steam supply. The 
number of oil feeds required is 
exactly the same as with a 
hydrostatic lubricator. With- 
out the atomizer box, piston 

_ _ rt . , valves require two oil feeds, one 

FIG. 152. Atomizer box. , 7 

for each end; but with the 

atomizer box, one feed for the center or one feed divided for each 
end, as the case may be, will suffice. 

The combination of a mechanical lubricator with a suitable 
atomizer box, in the author's opinion, offers the chief advantages 
of the best types of hydrostatic lubricators with all the advan- 
tages of mechanical lubricators. Only those oil feeds requiring 
to be atomized are carried to the atomizer box. Oil feeds for 
feeding oil under pressure to the axle boxes may be taken from the 
lubricator, and, if need be, the lubricator can be made with two 
compartments, so that a separate oil axle oil can be used for 
the bearings, and cylinder oil for the valves and cylinders. A 
hydrostatic lubricator can, of course, not be arranged to feed 
pressure oil to the axle boxes. 

Mechanical lubricators are fitted either in the cab near the 
driver or on the framing near the main points of lubrication. 



Motion may be taken from the back axle or from one of the 
rods, as shown in Figs. 153 and 154. 

The check valves should be designed to avoid steam's leaking 
back, and the vibration calls for special care; ordinary miter- 
seated valves are not satisfactory. Figure 155 shows one 
designed by the author, which has proved efficient under trying 
conditions. It will be seen that the spring operating the valve is 
on the oil side and not exposed to the steam; the valve has to be 
lifted until the cylindrical part is above the seat before the oil 
will be discharged. 


FIG. 153. Back-axle motion for mechanical lubricator. 

When oil is not pumped through the valve, the cylindrical 
portion below the head forms an effective seal against the 
entrance of steam into the oil pipe. By unscrewing the cleaning 
and testing plug, a straight passage is disclosed for cleaning the 
oil passage leading into the valve or cylinder. This plug is 
screwed back when testing the oil feeds. There are three oil 
holes below the head, through which the oil will exude. 

A similar type of check valve should also be used in the oil 
pipes from a mechanical lubricator to the axle boxes. The 
check valves should be fitted in accessible positions and as near 
the axle boxes as possible. 



Mechanical lubricators for locomotives, particularly when 
placed on the framing, should be provided with a steam-heating 

FIG. 154. Rod motion for mechanical lubricator. 

arrangement, as, if the cylinder oil becomes very thick or con- 
geals, the oil feeds will be considerably reduced or stop altogether. 

Section of Valve Stem 


FIG. 155. Locomotive check valve. 

As to placing the mechanical lubricators, they are undoubtedly 
best placed in the cab, where the lubricator is under the eye of the 
driver and stoker, and where each feed to each part of the engine 


can be properly controlled and regulated. This also makes it 
possible to give extra oil when required by having a flushing 
handle on the lubricator, by means of which all the oil feeds can 
be flushed. Where mechanical lubricators are placed on the 
frame, the driver cannot control and watch the feeds from the cab. 
If one of the feeds gets out of order, he will not be able to recognize 
this before the engine gives audible notice by grunting or otherwise, 
and then a great deal of damage may already have been done. 

It is felt by some engineers that the drivers should not be 
allowed to adjust the feeds when once set by an expert in the 
running shed or during a couple of days' service on the road. 
The lubricators can, of course, be arranged with locked adjust- 
ments, but the drivers should in any case be enabled to watch 
the sight-feed glasses and test the oil feeds ; they should also have 
access to the suction valves and to the flushing arrangement. 

The combination of a mechanically operated lubricator with 
an atomizer box appears to be the best solution for lubrication of 
all locomotives in those outlying countries which employ native 
drivers, as it is desirable that the lubrication be as automatic 
and foolproof as possible, and the control largely taken out of the 
driver's hands. 

For those countries in Europe and America where intelligent 
drivers are available, the hydrostatic lubricator, with intelligent 
care, is capable of giving good service, and it will probably con- 
tinue to be much used for saturated-steam conditions. As, 
however, the consumption of oil with mechanical lubricators can 
be automatically kept nearer the actual requirements than with 
the hydrostatic lubricator, which requires frequent and intelli- 
gent adjustment by the driver, it would not be surprising to find 
the mechanical lubricator gaining in favor for saturated-steam 
service. For superheated steam conditions the author thinks 
that the development will certainly be in favor of the mechanical 
lubricator, due attention being paid to the atomization principle. 


Most locomotives operate with rather high steam pressures, 
ranging .from 140 to 225 Ib. per square inch or even higher. 

Most works locomotives have slide valves, but many, railway 
locomotives have piston valves. Slide valves have been used 
with a moderate degree of superheat; but for high superheat, 


piston valves are universally, adopted, and in most cases also tail 
rods. Piston rings and pistons wear much better when tail rods 
are fitted. It is not considered advisable to exceed a steam tem- 
perature of 650F. 

In the early days much trouble was caused by the growth of 
cast iron when exposed to superheat temperatures. For a long 
time it was thought that the cylinder oil was to blame for the 
excessive wear and the many cracked cylinders, etc., but even- 
tually the swelling was found due to the combined carbon in the 
iron. A more suitable cast iron was discovered and solved the 
difficulty, and many railways found that good-quality filtered 
cylinder oils, which they had previously used with saturated 
steam, served quite well also with superheated steam. 

Owing to the wet-steam conditions often met with in locomo- 
tives or to bad water or to the necessity of keeping an extra-high 
water level before negotiating a long uphill gradient, experi- 
ence had already taught some railways that well-filtered green 
cylinder oils, compounded with from 6 to 10 per cent of acidless 
tallow oil, gave cleaner and better lubrication on a much reduced 
feed as compared with dark cylinder oils, whether straight mineral 
or compounded. The majority of railways, however, still use 
dark cylinder oils for all conditions, because they are lower in 
price than filtered cylinder oils. 

Experience has proved that locomotive-cylinder oils should 
certainly be compounded. If the conditions as regards priming, 
drawing in soot, etc., are not too trying, dark compounded cylin- 
der oils will give a reasonable amount of satisfaction; but under 
unfavorable conditions, compounded filtered cylinder oils should 
always be preferred, as they maintain the valves and cylinders 
in a much cleaner condition, which is worth a great deal from both 
a frictional and a wear-and-tear point of view. 

For works locomotives fitted with poor lubricators, it is usually 
a waste of money to use filtered cylinder oils, and dark com- 
pounded oils are recommended. The use of tallow should be 
discouraged, but it will often be found that collieries and steel- 
works buy low-priced, straight mineral, dark cylinder oils; that 
the locomotives use the oil extravagantly; and yet that the 
lubrication is so poor that engine drivers get tallow (or, if they 
are not allowed to have it, get it all the same) to keep their 
engines quiet. 


The bad effects of using tallow are mentioned on page 406. 
Locomotive-cylinder oils should obviously have good setting 
points, so that low-setting-point filtered cylinder oils should be 
recommended which will flow freely in the lubricators and give a 
uniform feed. Compounded filtered cylinder oils will also lubri- 
cate the air-pump cylinder satisfactorily, if fed sparingly, and 
very little oil vapor will be carried over with the compressed 
air into the air-brake system. 

The consumption of cylinder oil required for full lubrication 
varies from J to \Y pt. per 100 miles, according to the size of 
the locomotive. The oil consumption for the air pump varies 
between % and J^ pt. per 100 miles and should be kept as low as 

Where an engine has a long continuous run to make, it is good 
policy for one shift of driver and fireman to hand it over to the 
next shift with the lubricator fitted with oil; in this way, control 
of the various drivers' oil consumption is made quite easy. 


Locomotives* Cylinder Oilf 


Small 2 D.M.C. 

Larger 3 D.M.C. 


Saturated steam 3 F.M.C. or 3 D.M.C. 

Superheated steam 3 F.M.C. 

* For very wet steam, the same grades are recommended but with 10 per cent of compound, 
f For information as to the grades of cylinder oils recommended see p. 408. 



Compressed Air. Compressed air is used for a variety of 
purposes for supplying blowing air to blast furnaces and Bes- 
semer converters; for operating pneumatic tools, such as pneu- 
matic hammers, drills, riveters, etc., as used in engineering 
works, boiler shops, foundries, forge shops, shipyards, docks, 
and bridge building; for rock drills used in mines and quarries; 
for operating underground machinery in collieries; and for sink- 
ing tunnels and shafts. It is also used for operating different 
types of lifting and hoisting gear, railway-car brakes, electro- 
pneumatic signals, and pneumatic-tube carrying service ; for 
pumping water; for lifting and conveying liquids in breweries, 
distilleries, and chemical works; for aerating oils in large edible- 
oil refineries; and for spraying paint. 

Compressed air is employed for starting gas engines and other 
internal-combustion engines; also for injecting and atomizing 
fuel oil under furnaces or in Diesel engines. Very highly com- 
pressed air is used for producing oxygen and liquid air. 


Blowing engines supply large volumes of air at low pressure. 
Blast furnaces require air at 10 to 25 Ib. per square inch; Bessemer 
converters require it at 20 to 30 Ib. per square inch. 

Blowing engines operate at low speeds from 30 to 70 r.p.m. 
and are single-stage machines ; they are operated by either steam 
or gas engines; the gas engines are nearly always horizontal two- 
stroke cycle engines, driving the air cylinder tandem fashion. 
When driving the blowing engines by steam engines, the steam 
and air cylinders are also usually placed in tandem. In hori- 
zontal blowing engines the piston nearly always has a tail rod. 
When the tail-rod support is not present, the whole of the weight 
of the piston is sliding on the bottom of the cylinder, demanding 
the use of heavy-bodied oils. 




Air compressors compress air to high pressures. Colliery air 
compressors compress large volumes of air to a pressure of 60 to 
80 Ib. per square inch; they are sometimes single-stage compres- 
sors, but more frequently they are two-stage. 

The majority of compressors used for a variety of purposes, 
as enumerated above, compress air to a pressure of from 80 to 
120 Ib. per square inch. Small compressors operating at a high 
speed are frequently single-stage machines up to a delivery 
.pressure of 120 Ib. per square inch. Large compressors are nearly 


Blowing engines and air 

Air pressure, 
pounds per 
square inch 


Single or double 

Blowing engines: 
Blast furnace . 

10 to 25 I 

Bessemer converters 

20 to 30 ) 

30 to 70 

Double acting 

Air compressors (exclusive of Diesel-engine compressors) 
Small vertical compressors: 

Single stage 

Two stage 

Small horizontal compressors 

Single stage 

Two stage 

Large vertical compressors: 

Single stage 

Two stage 

Large horizontal compressors 

Single stage 

Two stage 

Up to 120 
Up to 450 

Up to 120 
Up to 450 

Up to 70 
Up to 150 

Up to 70 
Up to 150 

300 to 500 

150 to 250 

60 to 360 

40 to 150 

Single acting 

Double acting 

Usually single acting 

Double acting 

always two-stage machines when the air pressure exceeds 70 Ib. 
per square inch. Small- or medium-size compressors used in 
connection with semi-Diesel oil engines compress air to about 
400 to 450 Ib. per square inch and are two-stage machines. 

Air compressors used in connection with Diesel engines com- 
press air to a pressure of about 1,000 Ib. per square inch (see 
" Diesel Engines," page 564). 

Air compressors when used in connection with the production 
of oxygen compress air to a pressure of 2,000 Ib. per square 
inch and are usually four-stage machines. The types used for 


charging torpedoes compress air to 3,000 Ib. per square inch and 
are usually four- or five-stage machines. 

Horizontal air compressors are usually steam driven with steam 
and air cylinders in tandem. Vertical air compressors may be 
driven by steam, by an electric motor, or by belt from a trans- 
mission shaft. 

Blowing engines and air compressors may be classified as 
shown in the table on page 423. A compressor, whether it be a 
single- or a two-stage machine, is classified as small or large,, 
according to whether the volume of free air entering the machine 
is less or more than 1,000 cu. ft. per minute. 


Cooling. As blowing engines compress the air only to low 
pressure, the amount of heat produced is not very great, so that 
blowing-engine cylinders are practically never water cooled. 
In air compressors which compress the air to higher pressures 
and which operate at much higher speeds, the heat of compres- 
sion is great, particularly around the outlet valves, through which 
the hot compressed air is discharged. 

Cooling of the air-compressor cylinder therefore becomes 
necessary, and, under severe conditions, attempts are frequently 
made to cool also the parts in close proximity to the outlet valves. 
Without adequate cooling, the temperature would rise, causing 
unequal expansion and distortion of the compressor cylinder, 
valves, and valve seats. The lubricating-oil film between the 
piston rings and cylinder walls would be thinned out, losing its 
sealing power, and the compressed air would leak past the piston. 
The discharge valves would not keep airtight (distortion due to 
heat), resulting in wiredrawing and recompression of the air, 
charring of the lubricating oil, excessive carbonization, friction, 
and wear. 

If air at a temperature of 60F. is compressed in a one-stage 
compressor to 100 Ib. per square inch, its temperature will 
theoretically increase to 485F. ; under actual working conditions 
it will, however, be lower, owing to the cooling effect of the cool- 
ing-water jacket. 

When air is compressed at a temperature of 60F. to 100 Ib. 
pressure in a two-stage compressor, compressing the air to, say, 
35 Ib. pressure per square inch in the low-pressure cylinder and 



cooling it in an intercooler, the temperature of the air leaving the 
high-pressure cylinder will be considerably lower from 200 to 
250F. only in rare cases going as high as 300F. This example 
shows the value, as far as lubrication is concerned, of compressing 
air in several stages when the final air pressure required is high. 

The effect of the lower temperature is also that it takes con- 
siderably less power to compress the air (20 per cent less in the 
case just mentioned), this forming another strong reason in favor 
of multiple-stage compressors. 

The air is frequently cooled in an aftercooler when leaving the 
compressor. In cooling, it will deposit its surplus moisture 

1. Air inlet. 

2. Air outlet. 

3. Water level blow off cock. 

4. Bottom blow off cock. 

5. Water feed pipe connection. 

FIG. 156. Air purifier. 

and a large portion of the oil, which is thus prevented from reach- 
ing the receiver. 

Occasionally, a separator partly filled with water is fitted in 
series with or in place of the aftercooler (Fig. 156). The water 
assists in extracting dust and excess water from the air. A feed 
pipe and blowoff cock are fitted, as indicated, so that the water 
can be changed under pressure. Accumulated oil can be blown 
out from time to time through a scum cock. This may also be 
connected to an automatic trap. 

Filtration. Where the air is charged with dust, a strainer or 
filter should be fitted. It may be made of screens of wire gauze 
and may contain cotton wool or fiber, in order to retain the 


impurities. If the air is dirty, and impurities reach the compres- 
sor, the impurities will adhere and cling to the oil film, baking 
together into carbonaceous deposits. The intake air should 
therefore be taken from outside the compressor room and from 
as clean a place as possible. It may be freed from dust by passing 
through a container filled with 3-in. stones, coated with thick 
refuse oil and closed with grids to keep in the stones. The con- 
tainer and stones should be cleaned once or twice a year, and the 
stones recoated with oil. 


Feeding Oil into Air Intake. In small and medium-size air 
compressors, oil is occasionally introduced into the flow of air 
passing through the air-inlet pipe. The air atomizes and carries 
the oil in the form of a fine spray into the cylinder. The oil is 
cold, and the air is not a good carrying medium for oil, so that 
frequently this practice does not give the best results. 

In horizontal air compressors or blowing engines, if the oil is 
introduced into the air intake, it will with difficulty reach the 
top portion of the piston, as it arrives there only by slowly 
wedging its way up around the sides. This practice is therefore 
uneconomical, as a large quantity of oil has to be fed in order to 
ensure its reaching the top of the piston. 

In vertical air compressors the practice of feeding oil into the 
air-inlet pipe has a greater chance of distributing the oil than 
in horizontal air compressors, but it is also here rather wasteful 
and not conducive to the best results. 

Feeding Oil Direct. Generally speaking, it is better to feed the 
oil direct to the frictional surfaces, feeding it sparingly and uni- 
formly. In horizontal blowing engine or air-compressor cylin- 
ders, the oil is introduced at the center of the cylinder, either at 
one point, at the top; or at three points, one at the top and two 
lower down. It will then gradually work its way around the 
piston and form a complete sealing and lubricating film. 

In vertical cylinders, oil is introduced at two points, front and 
back, or at several points evenly spaced around the cylinder. 
Each oil inlet to the cylinder should preferably be fed by a sepa- 
rate oil pump, so that each feed can be controlled with certainty. 
If one oil pump supplies several oil inlets to the cylinder, the oil 
will take the path of least resistance and will not feed through 
those inlets which have become choked with dirt or deposit. 


Splash from Crank Chamber. In vertical enclosed high-speed 
air compressors where the external moving parts are enclosed in a 
crank chamber and lubricated by means of either the splash 
system of lubrication or the force-feed circulation system, the 
oil is either splashed or forced to all parts requiring lubrication, 
so that no separate oiling of the piston is required. On the 
contrary, the difficulty is usually to prevent too much oil from 
passing the piston rings and getting to the top of the piston, 
where, exposed to the high temperature and oxidizing effect of 
the air, it will in time bake into a carbonaceous deposit. 

The presence of a large amount of oil in the air also produces a 
similar deposit on the discharge valve, frequently causing great 

Valve Lubrication. Grid valves have large sliding surfaces 
which must be lubricated direct, by introducing the oil at several 
points, sparingly and uniformly, the oil gradually finding its way 
all over the sliding surfaces. 

Flap valves have hinges which must be oiled, sparingly and 
uniformly, the oil being introduced through feed pipes passing 
through the cylinder head. 

Leather-disk valves need no lubrication, but the leathers must 
be kept flexible and in good order by occasional application of 
neat's-foot or lard oil. 

Corliss valves (used only as suction valves) need lubrication, 
particularly at their ends, where the valves have their bearing 
surfaces; the oil must be introduced direct to these ends, spar- 
ingly and uniformly. The practice of fitting grease cups supply- 
ing grease to the valve ends is not to be recommended, partly 
because grease spreads only with difficulty over the rubbing sur- 
faces, and partly because it bakes together with the impurities in 
the intake air into a pasty, sludgy deposit, causing excessive 
friction and wear; some of the grease will reach the valve chamber 
and even the cylinder, where it will bake together with impurities 
and cause an objectionable varnish-like deposit. 

Poppet valves usually get sufficient lubrication from the oil in 
the air. 

Plate or disk valves require no lubrication. 

Bucket valves themselves require no lubrication, but their 
spindles must be sparingly lubricated. 

Lubrication of external parts is by means of splash oiling or 
force-feed circulation in the case of all high-speed enclosed-type 



air compressors; in open-type air compressors any of the many 
systems employed for bearing lubrication may be employed and 
do not call for any special comments. 

With splash oiling it is very important that the correct oil 
level be maintained, so that an adjustable overflow should prefer- 
ably be fitted to the crank chamber. 

Owing to the high efficiency of the force-feed circulation oiling 
system and to the vertical construction, vertical air compressors 
may operate at much higher speeds than horizontal air com- 
pressors, as indicated in the table on page 

Care should be taken that the piston rings 
and oil scrapers on the lower part of the 
trunk pistons are pegged and in good order; 
they will then wear to a fit with the cylinder 
and keep oiltight and compression-tight. 

In vertical enclosed-type air compressors 
employing the force-feed circulation oiling 
system, the oil pressure should not exceed 5 to 
15 Ib. With excessive oil pressure too much 
oil spray is formed, and too much oil is inclined 
to pass the pistons, particularly when the gov- 
ernor operates by throttling the intake air as the high vacuum 
created in the cylinders tends to draw the oil past the piston rings. 
Splash guards fitted over the crank webs and pegging the piston 
rings will assist materially in reducing the oil consumption. 


Usually, sight-feed drop oilers or mechanically operated force- 
feed lubricators are employed. 

Sight-feed drop oilers are subject to considerable variation in 
oil feed. If the containers are full, they will feed, say, 3 drops 
per minute; when they are nearly empty, they will feed, say, 1 
drop per minute. They also vary with the temperature of the 
oil, the feed increasing when the oil gets warm and thin ; in addi- 
tion, when they are adjusted to feed a very small amount of oil, 
which is required in air-compressor practice, grit or dirt may 
easily choke the needle valve controlling the oil feed. 

When a sight-feed drop oiler is to feed oil direct into the cylin- 
der, it must be enclosed, so that it will feed notwithstanding 

FIG. 157. 


the varying back pressure (see Fig. 157). A pressure-equalizing 
pipe connects the sight-feed chamber with the space above the 
oil in the oil container. 

The oil should preferably be fed by means of a reliable mechan- 
ically operated lubricator, having positive visible oil feeds and of 
such construction that it will feed the minimum quantity of oil 
with the greatest regularity and precision. The oil feeds, once 
adjusted, should remain absolutely constant, independent of the 
oil level in the container and independent of the viscosity of 
the oil. 


All open-type compressors are so constructed that an oil spe- 
cially chosen to suit the air-compressor requirements can be 
employed and applied quite independently of that used for the 
external moving parts. In enclosed-type air compressors the 
same oil must be used for air cylinders and bearings, and both 
requirements must be given consideration. The chief trouble 
in air-compressor lubrication is the formation of carbon deposits 
which may or may not bring about explosions or fires. 

Deposits. Deposits may form on the pistons, piston rings, and 
valves and in the discharge chambers, pipes, coolers, and receivers. 

Deposit on the piston rings may fill up the grooves and make 
them inoperative, causing heavy friction and wear and air leak- 
age past the piston. 

Deposit on the discharge valves and valve seats prevents the 
valves from seating properly; the hot compressed air will leak 
back into the cylinder on the suction stroke; recompression 
will cause the temperature of the discharge air to increase above 

If a discharge valve sticks in a partly open position, the air 
is wiredrawn and recompressed continuously; the hot air heats 
the valve, and the temperature may easily rise to 700F. or more, 
which is the spontaneous-ignition temperature of average-quality 
oil. The deposit now becomes incandescent, and accumulated 
oil will vaporize and burn or explode. Most explosions in colliery 
compressors appear to be caused by discharge valves' sticking. 

Deposit on the suction valve causes leakage on the compression 
stroke, and wiredrawing of the air causes heating of the valve 
and seat. 


Deposits in the discharge pipe restrict the opening; cases have 
been known where they have been almost choked, causing abnor- 
mally high pressure and temperature of the discharge air. 

Deposits may develop due to impurities in the intake air, inefficient 
cooling, too warm intake air, too much oil, or unsuitable oil. 

Impurities in Intake Air. When air compressors operate in 
dusty surroundings, as in collieries and quarries, the dust fre- 
quently brings about deposits inside the compressor cylinders, 
valves, etc., unless the intake air is filtered. 

In one colliery several explosions had occurred in one of their 
compressors; but when it was arranged to filter the intake air 
(which revealed how very dirty the air was), no further explosions 
took place. 

In another colliery an electrically driven compressor was 
placed down a pit in a place where the coal trains passed by, 
with the result that the pistons and valves were constantly 
choking up with deposit, and heavy wear took place. A sample 
of deposit taken from the valves showed the following analysis: 

Per Cent 

Moisture Traces 

Oil 26.0 

Volatile matter (coal dust and oil carbon) 54.0 

Fixed carbon and silica 0.9 

Iron oxide (chiefly wear) 18.1 

Balance undetermined 1.1 


A filter was then installed, and the compressor kept very much 

Inefficient Cooling may be due to furring up of the water 
jackets; the result is that the oil is charred and bakes together 
with metallic wearings from the piston, piston rings, and cylinder. 

Neglect on the part of the attendant in not turning on the 
cooling-water supply when starting the compressor has been 
responsible for such deposits and even for explosions. 

Warm Intake Air. The warmer the intake air the hotter will 
be the discharge air, the results being similar to those of inefficient 
cooling. A certain difference in temperature of the intake air 
means a much bigger difference in that of the discharge air, which 
emphasizes the desirability of having the intake air as cool as 


Too Much Oil. Air compressors require very little oil for 
lubrication because the oil remains a long time once inside the 
compressor; there is no steam to wash it away as in steam 
engines, and there are no high temperatures to burn it away as in 
internal-combustion engines. 

Air compressors can rarely get little enough oil; the excess oil 
remaining on the piston or valves often gets charred into a hard 
carbonaceous deposit. 

Unsuitable Oil. The character of the oil itself greatly influ- 
ences the character and amount of carbon deposit formed. 

Pale oils containing chiefly saturated hydrocarbons naph- 
thenes or paraffins produce less oil carbon than such dark- 
colored oils which contain types of hydrocarbons easily 
decomposed by oxidation. 

Distilled oils produce much less deposit than undistilled oils. 
Exposed to high temperatures they distill away almost com- 
pletely, leaving comparatively little residue behind, whereas 
Undistilled oils exposed to high temperatures distill only partly, 
leaving a spongy carbonaceous residue behind. Dark cylinder 
oils leave much more residue than filtered cylinder oils and ought 
never to be used for air-compressor service. 

As regards fixed oil, it is obvious that semidrying or drying 
oils cannot be permitted as an ingredient in air-compressor oils, 
but the presence of a small percentage say, 3 per cent of non- 
drying animal oil is not detrimental to air-compressor lubrication; 
in fact, it has proved a distinct advantage in multiple-stage high- 
pressure air compressors where the air in the higher stages is wet 
(see " Diesel Compressors/ 7 page 564). For low- or moderate- 
pressure compressors, when the air is comparatively dry, the 
admixture of fixed oil is unnecessary. 

Oils too heavy in viscosity are largely responsible for deposits; 
the dust and dirt in the air adhere to the sluggish oil and form a 
black pasty deposit. 

The cry for high-flash-point compressor oils, which comes up now 
and again after compressor explosions in mines, usually meets with 
a far too ready response. High flash point means high viscosity 
(large percentage of filtered cylinder stock in the oil), and this 
inevitably means more trouble with carbon deposit than ever. 

In colliery compressors using air-compressor oils with a flash 
point of over 500F. (steam-cylinder oils) the coal dust bakes 


together with the oil and presents a smooth glossy surface, due 
to the pitch and tar contained in the coal dust. Such high-flash- 
point oils have one virtue, however, in that they do not give off 
much vapor exposed to the normal air temperatures in an air 
compressor. Their use is therefore justified in fact, may be 
quite necessary where lower-flash-point oils give off so much 
vapor that they affect the throats and lungs of the workmen 
in tunnel work, collieries below ground, air-worked machinery 
in confined spaces, etc. For such conditions, reasonably low- 
viscosity filtered cylinder oils should be employed. The flash 
point is no safe criterion as to the amount of vapor given off below 
the flash point. Speaking generally, high-viscosity oils act 
sluggishly and are inclined to retain much of the dust, particularly 
on the discharge valves, where the maximum temperature exists. 
When such oils are used, and the air is dirty, it must be filtered, 
and the compressor pipes and receivers should be frequently 
examined and cleaned, so that, notwithstanding the sluggish oil, 
the danger of explosions may be avoided. 

Low-viscosity oils assist in maintaining the compressor in a 
clean condition, notwithstanding dirty surroundings; the dirt 
that gets in i kept moving and is largely passed through the com- 
pressor and out of the discharge valve into the discharge pipe, 
aftercooler, and receiver. 

Soap and water are excellent for cleaning purposes, but their 
use as a lubricant does not dissolve existing deposits; in fact, 
more deposit is formed, as the water evaporates. In one case, a 
2-in. deposit (which ignited at 400F.) was formed inside the 
discharge pipe of a compressor, lubricated entirely by soap and 
water. Explosions have been reported to have occurred when 
soap and water have been used exclusively for lubrication, but 
the author has no personal knowledge of any such cases. 


We have seen several reasons for the production of abnormally 
high temperatures. The heat emanates chiefly from the dis- 
charge valve or valves, and it is probably safe to say that fires 
or explosions originate at or near the discharge-valve chamber. 

Exposed to high temperature, the accumulated oil or oily 
deposit will begin to emit vapor at 120 to 150F. below the open 


flash point of the oil. As the temperature increases, the oil will 
vaporize more vigorously; and when the temperature is well 
above the flash point, the mixture of oil vapor and air may 
easily accumulate in or near the discharge-valve chamber and 
be in the right condition to explode. Perhaps a small piece of 
deposit on the discharge valve begins to glow sufficiently to fire 
the mixture. A temperature of about 700F. is sufficient to 
ignite the oil vapors spontaneously, and a fire or explosion follows. 

Experience seems to show that in large moderate-pressure 
compressors explosions do not occur if the intake air is filtered 
or if deposits are not allowed to accumulate in too great quanti- 
ties. When there are no deposits there can be no fire, therefore 
no explosions. The amount of oil used for lubrication in large 
compressors is so small compared with the large volume of air 
passing through the compressor that the oil vapors formed, 
even under high-temperature conditions, are so diluted that they 
cannot explode. If an explosion occurs, it is frequently too 
weak to burst pipings or receivers. 

The high temperature may, of course, ignite accumulated oily 
deposits in the discharge pipe, in which case the fire will spread 
slowly to the receivers. The burning deposit may make the 
receiver walls red-hot, so that they burst, being unable to with- 
stand the normal receiver pressure. 

In one typical case of a colliery compressor the accumulation 
of coal dust and oil in pipes and receiver had not been cleaned 
out for 2 years; there was a weak explosion, and the deposit 
burned for a considerable time, causing men in the pit operating 
coal cutters to cease work owing to the obnoxious fumes in the 
compressed air. 

In another case, a leaking joint on the discharge pipe close to 
the compressor had been "cured" by driving a piece of wood into 
the joint. The point of the wood protruded inside the pipe and 
ignited spontaneously, owing to abnormally hot discharge air. 
The fire spread to the receiver, and, the latter being opened up, 
3 cwt. of deposit accumulated over 7 years was removed or, 
rather, what remained after most of the combustible matter had 
burned aw;ay. 

If the dust, which together with the oil forms the deposit, is 
itself inflammable, such as coal dust, the danger of the deposit's 
taking fire is, of course, greater than where it consists of non- 


inflammable ingredients, such as fine sand and dust in quarries 
and iron mines. 

In multiple-stage high-pressure compressors, where the volume 
of air discharged is comparatively small, the amount of oil used 
for lubrication and intermingled with the air is appreciable; 
and under conditions of abnormal temperatures, explosive mix- 
tures of oil vapor and air are formed, which will bring about 
violent explosions, when the spontaneous-ignition temperature 
is reached. Such explosions may occur even if the amount of 
accumulated deposit is small. 

Afterburning of deposit, which is a characteristic feature of 
most " explosions " in large moderate-pressure compressors, does 
not occur in high-pressure compressors. If an explosion occurs 
in the very confined spaces, it is very violent and usually shatters 
the piping, receiver, etc. 

NOTE;. Valve pockets or discharge chambers and pipes should be so 
designed that there are no cavities where mixtures of oil vapor and air may 
remain stagnant. 

Spontaneous -ignition Temperatures. The temperature at 
which oil vapor and air ignite spontaneously, i.e., without the 
aid of a spark, is higher the lower the viscosity of the oil. Speaking 
generally, the more complex and the more viscous a petroleum 
product is the lower is its spontaneous-ignition temperature. 
For example, kerosene ignites spontaneously in air at a lower 
temperature than gasoline. The compression in kerosene-oil 
engines is lower than in gasoline engines for this very reason, as 
the danger of preignition is greater with kerosene. 
. It will, therefore, be realized that the danger of explosions is 
not lessened by the use of very high-flash-point oils. Quite apart 
from the fact that such oils are extremely viscous and favor forma- 
tion of deposits, the mixture of air and vaporized oil is spontane- 
ously ignited at lower temperatures than with a lower viscosity oil. 
It might be asked, Why, then, not go to the other extreme and use 
very low-flash oils? Up to a certain point this view is certainly 
justified and correct. But with extremely volatile oils, although 
they will tend to keep the internal conditions clean and thus 
minimize danger of explosion, yet they vaporize so much exposed 
to normal compressor temperatures that the presence of vapors in 
the compressed air will become troublesome, and, furthermore, 
such oils will not satisfy the requirements as regards lubrication. 


Too thin oils will not seal the pistons and will cause excessive 
internal friction and wear. 

In view of what is said above, its seems probable that very few 
explosions have been caused on the discharge side of a compressor 
by injecting kerosene into the compressor for cleaning purposes; 
but when kerosene explosions have occurred they have usually 
been in the compressor cylinder itself, the ignition taking place 
through the suction valves on the approach of a naked light. 
For the same reason, no naked light should be used when opening 
up receivers or inter coolers for examination. 

The following case shows, however, that the flame caused by 
the presence of kerosene may be carried right through the 
compressor and ignite a mixture of air and oil vapor on the 
discharge side. 

In a compressor, in which the valves had been reseated and the 
cylinder cleaned out, the cleaning was done with kerosene. When the 
compressor was started up, the engine attendant came to the conclusion 
that something was wrong with one of the suction valves and took up 
a candle for the purpose of inspecting it. The result was an explosion, 
the discharge pipe being blown to pieces for a length of about 10 yd. It 
was evident that a quantity of kerosene was pocketed in the suction- 
valve chamber and that as the engine acquired the usual working tem- 
perature, after a short run, the heat was sufficient to vaporize the 
kerosene. When the engine attendant inspected the valve, the candle 
flame ignited the kerosene, the flame was carried through to the dis- 
charge pipe, and the explosion followed. 

Air-compressor Rules. The following rules should be observed 
in order to avoid danger of explosions: 

1. Intake air should be taken from outside the engine room and 
should be cold, clean, and, if necessary, filtered. 

2. A sparing and uniform amount of a carefully selected com- 
pressor oil should be supplied, with frequent drainage of inter- 
cooler and aftercooler for water and oil. 

3. Good cooling of the cylinder should be practiced, including 
discharge-valve chambers, as abnormal temperatures emanate 
from these valves. The cooling water must always be turned on 
before the air compressor is started. 

4. Temperatures should be taken regularly of intake and dis- 
charge air, as abnormal rise in temperature is a sure indication of 



5. An aftercooler should be fitted in the discharge line before 
the receiver under difficult conditions, so that only cold air enters 
the receiver. 

6. Compressor pressure gauges should be periodically examined 
and corrected by comparison with standard gauges. 

7. There should be frequent inspection and cleaning of water 
jackets, valves, discharge pipe, aftercooler, and receiver; in 
multiple-stage air compressors, discharge valves should be 
examined every week; low-pressure valves, every month; receiver 
and coolers, every month to every 6 months, depending upon the 

8. Kerosene should never be used for cleaning the compressor 
or pipes internally, as it evaporates and forms an explosive mix- 
ture with the air. Soap and water should preferably be used for 
cleaning, the surfaces being afterward wiped clean and oiled with 
compressor oil to prevent rusting while standing. 


Air-compressor oils, in view of what is said in the preceding 
chapter, should preferably be pale-colored straight-run distil- 
lates, highly refined and filtered, containing as few unsaturated 
hydrocarbons as possible. They should preferably contain little 
or no cylinder stock. 

Where, in order to obtain a heavy viscosity, the admixture of 
filtered cylinder stock becomes necessary, the distilled oil should 
be as viscous as possible so as to minimize the percentage of 
cylinder stock required in the finished oil. 

Air-compressor oils of four different viscosities are required to 
lubricate the cylinders and valves of all types of blowing engines 
and air compressors, as indicated in the table below. 




at 50C. in 

Flash point open, 

Flash point closed, 




















* 475 

* See table, p. 57. 

t Compressor oil 4 U a filtered steam-cylinder oil. 



These four oils are usually straight mineral oils; but for 
multiple-stage compressors, as Diesel compressors, oils 2 and 3 
are recommended and should preferably contain from 3 to 6 per 
cent of a nondrying, acid-free, fixed oil. 

The following chart gives specific recommendations for the 
various types of blowing engines and air compressors. 

For Blowing Engines and Air Compressors 

Type of blowing engine or oil compressor 

Number of 

Final air 
pounds per 
square inch 


Blowing engines: 

Blowing cylinder horizontal, no tail rod 

Single stage 

10 to 30 


Horizontal, with tail rod 

Single stage 

10 to 30 



Single stage 

10 to 30 

1 or 2 

Air compressors: 


Compressing less than 1,000 cu. ft. of free 

air per minute 

i Single stage 

Below 70 


Small compressors are usually enclosed and 

Two stage 

Below 150 


use the same oil for cylinders and bearings 

Single stage 

70 to 120 


Two stage 

150 to 450 



Compressing more than 1,000 cu. ft. of free 

air per minute 

Horizontal cylinders, no tail rod 

f Single stage 

Below 70 | 

3 or 4 

I Two stage 

Below 150 / 

Horizontal cylinders, with tail rod 

( Single stage 

Below 70 > 

2 or 3 

I Two stage 

Below 150 / 

For large compressors compressing above 

3 or 4 

the pressures given 

Large horizontal compressors are usually 

open type and use separate bearing oils 


Large vertical compressors are frequently en- 

closed type, employing force-feed circu- 

lation for the bearings, and the same oil is 

used throughout 

NOTE 1. Where a compressor is delivering air to air-worked engines placed in confined 
spaces (tunnel work, etc.), use a heavier viscosity (less volatile) oil than the one indicated in 
the chart. 


Dry-air pumps, or vacuum pumps, e.g., as employed in con- 
denser plants for steam engines or steam turbines, are a kind of 


air compressor; they compress the small amount of air leaking 
into the system and discharge it at atmospheric pressure. 

Dry-air pumps are often constructed with slide valves, and the 
lubrication of these valves is troublesome and difficult. The oil 
is subjected to the vacuum under conditions of high temperature, 
owing to the surfaces' being in touch with hot steam and to the 
additional heat created by valve friction. The result is that the 
oil is distilled " vacuum distilled" and is oxidized by the air, 
forming a sticky carbonaceous deposit. The remedy lies in 
using the oil with the utmost economy and applying it regularly 
and uniformly, preferably by means of a mechanically operated 
lubricator. The less oil consumed the less carbon is formed. An 
excellent idea is to introduce a jet of steam through a ^-in. 
exhaust-steam pipe taken from the steam engine driving the air 
pumps, e.g., in the Alberger pump. There must be no valves in 
this pipe; this admission of moist steam greatly minimizes the 
formation of carbon. 

Many engineers, when they have experienced trouble with a 
medium- viscosity compressor oil, jump to the conclusion that 
by using a higher flash-point oil the carbonization will be over- 
come; they therefore use steam-cylinder oils, "the thicker the 
better/ 7 and find the carbonization much worse than before, 
notwithstanding their endeavor to use as little as possible. As 
the oil is volatilized during use, it is obvious that a distilled lubri- 
cating oil, which has already been volatilized when it was being 
manufactured, must have less tendency to leave a residue than 
steam-cylinder oils which are undistilled. 

Experience proves that the best results are obtained by using 
compressor oil 2, as pale as possible, without cylinder stock and 
preferably slightly compounded so as to make it combine with 
the moisture, which is always present. 

NOTE 1. No. 4 compressor oil must be used only if there are very special 
reasons for using such a heavy-viscosity oil, e.g., the necessity of having an 
absolute minimum of oil vapor in the compressed air or bad mechanical 
conditions in large horizontal compressors. 

NOTE 2. Glycerin must be used for compressors in breweries, as even 
slight traces of mineral-oil vapor in the air will be absorbed by the beer and 
affect the taste, whereas glycerin has no detrimental effect whatever. 

NOTE 3. For three- and four-stage compressors, the same oils are recom- 
mended as for Diesel compressors (see page 572), viz., compressor oils 2 and 3 
compounded with 3 to 6 per cent of fixed oil. 


Similar conditions exist in a number of other vacuum pumps, 
e.g., those used in connection with sugar-evaporating pans. 


Compressed air is used for operating a variety of engines, 
machinery, and tools as indicated in the beginning of this section 
(page 422). 

Air-operated Engines. -The operating temperatures of the 
engines, etc., determine what viscosity oil is to be used. 

Air engines operating coal cutters are usually fairly warm and 
demand an oil like bearing oil 5 (see page 135). As the temper- 
atures are never more than moderate, there is no danger of car- 
bonization^ taking place, so that a bearing oil of suitable viscosity 
will do all that is required. As a rule, the operating temperatures 
are low, particularly when the engines or tools operate with air 
expansion, because the air becomes cold when it expands. 

Such low temperatures may bring about trouble by the lubri- 
cant's congealing or the engine's becoming choked with snow. 
The amount of moisture in the compressed air is often consider- 
able. When, for example, warm compressed air is sent down the 
shaft in a coal mine, it cools, and some of the moisture condenses ; 
if it is not efficiently drained out just before reaching the engine, 
it will freeze into snow, lodge in the exhaust port, and accumu- 
late till the engine pulls up. Even if the lubricating oil does not 
congeal, it will not clear the exhaust, but an admixture of glycer- 
in with the oil, say from 30 to 50 per cent, will usually thaw the 
snow and keep the exhaust clean. The mineral oil should have a 
cold test of, say, 25F. for such extreme cases, but usually a 
zero cold test will be found satisfactory. Large air-operated 
hammers for forging purposes should preferably have the oil 
introduced by means of a mechanical lubricator, the movement 
being taken from the hand lever (see Fig. 158). For such large 
hammers medium-bodied oils are preferable, as the operating 
temperatures are very moderate. 

A class of air-operated engine difficult to lubricate is the one 
in torpedoes. It may have three cylinders enclosed in a crank- 
case. The oil is forced into the main bearings, then through tiny 
holes in the crankshaft, say ^oo i n -> into the crankpins, while 
the pistons are lubricated by splash from the crankcase. The 
oily exhaust air from the engine may be used for lubricating some 



of the gears. The oil is forced into the bearings by means of air 
pressure. Toward the end of the run the air pressure drops, and 
the oil supply diminishes, as the resistance toward the oil flow 

Fio. 158. Air-operated hammer with mechanical lubricator. 

through the tiny passages remains unaltered. Simultaneously, 
the air is heated to maintain sufficient engine power; the hot air 
burns and oxidizes the oil in the cylinders. 

The conditions are therefore irregular oil feed, i.e., overfeeding 
most of the time, and exposure to high temperatures and air 



oxidation. All mineral oils produce too much carbon under 
these conditions. The oil that has given most satisfaction is 
cold-pressed, highly refined, acid-free neat's-foot oil or its equivalent. 
Such an oil has a very high flash point without being unduly 
viscous, and it gives practically clean lubrication. 

Pneumatic Tools. Pneumatic tools operate at very high 
speed (often several thousand strokes per minute), and the parts 
have exceedingly fine clearance. They are therefore very sensi- 
tive, and the air consumption may easily increase 25 per cent or 
more if too viscous oils are used. Oils for pneumatic tools should 

1 Oil Chamber 

2 Filling Plug 

3 Adjaiting Needlo 

4 Direction of Air 

FlG. 159. 

FIG. 160. 

Pneumatic tool oiler. 

therefore be very light-viscosity oils and have low sometimes 
very low cold tests to prevent them from congealing and clog- 
ging the tools. The oil is usually fed into the tool at intervals, 
say every hour. If a tool freezes up, an injection of glycerin 
will usually thaw the snow and clear the exhaust, after which the 
usual low-viscosity oil may again be applied. 

Several attempts have been made to introduce the oil sparingly 
and uniformly into the air before it reaches the tool, so as to avoid 
underlubrication. Figure 159 illustrates an oiler with a needle 
adjustment valve used by the Chicago Pneumatic Tool Company. 
Figure 160 shows an outside view; the direction of the air must 
be indicated. 

It has been found that delays caused by underlubrication and 
stoppage of tools are reduced as well as the cost of maintenance, 
when such oilers are used; the filling of the oil chambers can be 


done in the toolroom at night, when the tools are made ready for 
the following day's service. 

Great care must be taken to ensure that the air-supply piping 
and also exhaust piping (if the latter is fitted) are free from 
dirt and chips and that they are thoroughly blown out before 
final connection is made to the tool, so that no dust or foreign 
matter will be carried to the working parts, and the exhaust pipe 
will be clear. There is usually a strainer at the end of the branch 
air pipe to which the flexible hose is attached. This strainer is 


For Air-operated Engines and Pneumatic Tools 
Air-operated engines: 

Air-operated coal cutters Bearing oil 5 (see page 135) 

This oil also to be used for gen- 
eral lubrication of the coal 

Air-operated haulage engines, etc. Refrigerator oil 1 or 2 (seepage 460) 

or mixtures of these with up to 50 
per cent of glycerin where the 
exhaust is liable to choke with snow 
Large air-operated forging ham- Bearing oil 4 


Belt-driven pneumatic forging ham- Air-compressor oil 2 (see page 436) 
mers in which air is compressed 
and used as an air buffer 
Air engines in torpedoes Highly refined neat's-foot oil with 

a 0F. cold test 
Pneumatic tools: 

Large pneumatic drills, etc. Refrigerator oil 1 or 2 (see page 460) 

Smaller pneumatic tools Light pneumatic-tool oil* (see below) 

For gear cases in pneumatic tools Filtered cylinder oil with a poor cold 

test, say 80F. 

* Light Pneumatic Tool Oil: Pale, straight-run distillate, highly refined, having a Saybolt 
viscosity at 104F. of approximately 80 sec., and a setting point of 25 to -f-15F. according 
to the temperature conditions under which the tools operate. 

made of fine-mesh brass gauze or cloth and retains scale and 
impurities, which would otherwise injure the tools. Even a 
small piece of rubber of the air hose will put the tool out of action. 

It is good practice to immerse pneumatic tools in a bath of 
gasoline or kerosene overnight, then blow them out under pres- 
sure and oil them thoroughly before use. 

For lubricating the gears in many types of tools, a filtered, 
poor-cold-test, say 80F., filtered steam-cylinder oil will give 
good service; it will be semisolid at ordinary temperatures. This 


oil may be injected into the gearcase by a syringe, say every few 
working hours. 

It is important that the compressed-air pipe system be properly 
drained, to prevent water from getting into the tools, as such 
water would cause rusting of the pipes and also of the working 
parts in the tools, besides clogging the tools with snow, when they 
operate with air expansion. 



Refrigerating machines are used for producing cold, being 
employed in a great variety of installations, such as ice-manu- 
facturing plants, breweries, distilleries, dairies, sugar factories, 
chocolate factories, slaughterhouses, cold-storage plants, oil mills, 
margarine works, stearin works, paraffin works, chemical works 
of various kinds, artificial skating rinks, for domestic purposes 
in large houses or hotels, hospitals, etc., public mortuaries, mining 
operations (sinking shafts through wet sand), also in fishing 
vessels (freezing fish), food-transport ships, modern passenger 
ships, warships (cooling ammunition chambers), etc. 


A great variety of refrigerating machines are in use; they can, 
however, be classified according to the system of refrigeration 
employed as follows: 

Absorption machines. 
Compression machines. 

Absorption Machines. These machies usually operate with 
ammonia. They are manufactured only by a small number of 
firms. No lubrication is required except for the circulating 
pumps, the lubrication of which presents no difficulty. 

Compression Machines. In these machines the cooling 
medium the refrigerant at one stage of the process is com- 
pressed; hence the name compression machines. They are built 
in all sizes, requiring from ^ horsepower for the smallest units up 
to 800 horsepower for the largest units in large installations. 

According to the refrigerant employed, these machines may be 
divided into: 

Cold-air machines. 
Sulphurous acid machines. 
Ammonia machines. 
Carbonic acid machines. 



The refrigerants are, respectively: 


Sulphur dioxide (SO 2 ) 
Ammonia (NH 8 ) 
Carbon dioxide (CO 2 ) 

Cold-air machines are very bulky, and only a few machines 
are in existence. They were at one time used to some extent 
on board ship but have now been displaced by carbonic acid 
machines. They usually have two large cylinders. The air is 
compressed in one of these cylinders and expands and cools in 
the other. Glycerin is used for lubrication, as mineral oil gives 
the air a burnt odor, which taints meat. 

Sulphurous acid machines are bulky about two and one half 
times the size of ammonia machines and are now seldom used. 
As the sulphurous acid is a lubricant in itself, no internal lubrica- 
tion is required. 

Ammonia machines and carbonic acid machines are practically 
the only two types of refrigerating machines employed in modern 

Ammonia machines are generally employed in land installa- 
tions. They take less power to operate than carbonic acid 
machines, and the pressures carried in the system are considerably 
lower than the pressures in carbonic acid systems. 

The principal objections to ammonia machines are that ammo- 
nia leaking out from the system has an unpleasant penetrating 
odor and is suffocating; on the other hand, the odor makes a 
leakage easily noticeable. 

Carbonic acid machines are used almost exclusively on 
board ship; they take up considerably less room than ammo- 
nia machines. Carbonic acid is odorless; a leakage is therefore 
not easily detected, and good ventilating arrangements are 

During recent years, a variety of small high-speed refrigerating 
machines has been developed employing refrigerants such as 
methyl chloride (CH 3 C1), dichloroethylene (C 2 H 2 C1 2 ), dichloro- 
methane (CH 2 C1 2 ), and dichlorodifluormethane (CC1 2 F 2 ). 

Common to all these refrigerants is the fact that th^y dissolve 
mineral lubricating oils and therefore are not easily separated 
from the oil, as is the case with other refrigerants. 




Figure 161 illustrates the main elements found in all refrigerat- 
ing plants working on the compression system. The principle of 
operation, whether ammonia machines or carbonic acid machines 
are employed, is exactly the same, only the pressure and tempera- 
tures being different. 

The following description is given for ammonia machines, the 
particular in brackets referring to carbonic acid machines. 

The elements are the following: 


Oil separator. 


Regulating or expansion valve. 


Dirt catcher. 

The compressor (1) draws in gaseous ammonia from the suc- 
tion pipe (7), leading into the suction valve. The ammonia is 

I OompreMor 

8 Oil Seperator 

3 Condenser 

4 Expansion Valve 

5 Evaporator 

6 Dirt Catcher 

7 Suction Pipe 

8 Discharge Pipe 

9 Non return Valve 

10 Discharge Pipe to 

II Brine Tank 

12 Outgoing Brine Pipes 

13 Return Brine Pipes 

14 On Heater 


FIG. 161. Refrigeration system. 

compressed to a pressure of from 120 to 180 Ib. per square inch 
(CO 2 from 900 to 1,200 Ib. per square inch) and delivered at a 
temperature of 85 to 150F. (CO 2 160 to 170F.) through dis- 
charge pipe (8), through a nonreturn valve (9) into the oil separa- 
tor (2), from which it is conveyed through a pipe (10) into cooling 
coils in the condenser (3). 


Cold water passing through the condenser cools and liquifies 
the hot ammonia. The cold and liquified ammonia now passes 
through the regulating or expansion valve (4) into the coils of the 
evaporator (5). 

The pressure in the evaporator coils is low, from 15 to 45 Ib. per 
square inch (CO 2 from 200 to 400 Ib. per square inch). 

The effect of this considerable fall in pressure is that the liquid 
ammonia evaporates and in doing so cools down considerably 
below freezing point, the temperature being from 20F. to 
+ 15F. (CO 2 : -30F. to + 15F.). 

The cold evaporator coils are seldom placed directly where it is 
desired to produce cold. Usually they are placed in a tank (11), 
through which is circulated a nonf reezing brine (a salt solution) ; 
the brine, in passing over and around the cold evaporator coils, 
cools in contact with the coils. By means of a pump the cold 
brine can be pumped away through pipes (12) to the place where 
it is desired to produce cold. The brine returns through pipes 
(13) to the evaporator tank to be cooled again. 

The ammonia vapor leaves the evaporator coils at a tem- 
perature slightly lower than the temperature of the brine and 
returns through the dirt catcher (6) to the compressor, continu- 
ing the cycle of operations just described. 

During recent years, a new system of ammonia refrigeration, 
called the dry-compression system, has come into use. It oper- 
ates on the same principle as those machines already described, 
which are wet-compression machines, the chief difference being 
that the temperature of the ammonia in passing through the com- 
pressor is from 160 to 190F. higher than that in wet-compression 
machines. The heat developed in a dry-compression machine 
is so high that it becomes necessary to surround the compressor 
cylinder with a cooling-water jacket. 


Small compressors are driven by belt or rope drive. 

Large compressors are usually operated by a steam engine, the 
steam engine and the compressor having a common crankshaft. 
Sometimes the steam-engine cylinder is placed in tandeip with the 
compressor cylinder. All compressors operate at low' speed, as 
at high speed the operation of the valves becomes irregular. The 



compressors are built either vertical or horizontal, the practice in 
this respect varying in different countries. 

Most large compressors and many small ones are double acting, 
as the one illustrated in Fig. U 162; but frequently vertical com- 
pressors are single acting, even in large sizes, there being only 
one suction valve and one delivery valve. 

The cylinders of ammonia compressors are constructed chiefly 
of cast iron or steel, as copper or bronze parts would be attacked 
by ammonia. The cylinders of carbonic acid machines must be 
made very strong, on account of the high working pressure. 
They are generally made of a forged block of steel, suitably 
bored and finished. 


FIQ. 162. Ammonia compressor. 

Stuffing Box. The most important part of the compressor and 
the most difficult part to keep in good working order and well 
lubricated is the stuffing box. The object of the stuffing box 
is to prevent the escape of ammonia or carbonic acid from the 
cylinder and also to prevent outside air or moisture from entering 
the compressor through the stuffing box. 

Figure 162 illustrates one type of ammonia-compressor stuffing 
box. The bottom ring consists of white metal; the packing rings 
are of cotton, saturated with oil. (1) is the so-called "lantern" 
which has a hollow space filled with oil around the piston rod; 
then follow more cotton packing rings, and sometimes a rubber 
ring, all the packing rings being squeezed together by means of the 
stuffing-box gland (2). For sealing this gland, oil is introduced 
through the inlet (3), and overflows through the outlet (4). Oil 



is prevented from creeping out along the rod by means of the 
small stuffing box (5). 

The oil film on the piston rod absorbs ammonia vapors from 
the inside of the compressor. These ammonia vapors escape 
in the lantern and rise through the pipe (6), from which they 
are passed over into the suction pipe of the compressor, so that 
no refrigerant is lost. 

As the oil film swells on the piston rod inside the compressor, 
owing to absorption of ammonia or CO 2 , a portion of oil is 

FIG. 163. Stuffing box with leather packing. 

continuously scraped off by the gland on the outward motion of 
the piston rod, and this oil serves to lubricate the piston. 

As rubber is destroyed by the action of mineral oil and swells 
when absorbing CC>2, most manufacturers have discontinued its 
use in gland packings in favor of metallic or leather packing. 

Figure 163 shows one form of stuffing box for a CC>2 machine. 
It is very accurately machined, and in the bottom is introduced 
a bronze ring, after which three sets of bronze rings and leather 
rings are introduced, then the lantern to which the oil is applied 
under pressure, subsequently two sets of bronze rings antf leather 
rings followed by a ring of cotton; and then the stuffing-box 
gland keeps the whole packing in position. The bronze rings are 



all a good fit against the inside of the stuffing box but do not 
touch the piston rod. The leather packing rings are thinned out 
toward the cylinder so that they form an elastic edge, and the 
pressure in trying to escape from the cylinder automatically 
causes the leather rings to press against the piston rod and thus 
to prevent leakage. The life of the leather packing is usually one 
season, sometimes two. 

In some stuffing boxes there is an oil well near the front portion 
of the gland, covered with a lid. The gland should be so adjusted 

FIG. 164. Stuffing box with metallic packing. 

that occasional bubbles of CO2 are to be seen rising from this 
well, but there should not be sufficient C(>2 escaping to cause 
foaming. If there are no bubbles, the packing is too tight. 

Figure 164 illustrates a metallic packing built up of two spirals 
screwed into one another, the spiral (1) being of white metal and 
of triangular section; this gets squeezed tightly against the rod 
when the stuffing-box gland (3) is tightened up. The spiral (2) 
is also of triangular section but made of steel and forces itself 
toward the walls of the stuffing box, leaving spaces near the rod 
where the lubricating oil can accumulate for the purpose of 
lubricating the piston rod. The small stuffing box serves the 
same purpose as in Fig. 162. When this gland is in good condi- 



tion and properly adjusted, only very little oil reaches the 
interior of the compressor. 

In dry-compression machines, packings containing cotton, 
leather, or rubber cannot be employed, as they will be destroyed 
by the heat. Metallic packings are used, as the one illustrated in 
Fig. 164. Another type of metallic packing used for dry-com- 
pression machines consists of a num- 
ber of rings, each in two halves sur- 
rounding the piston rod and pressed 
lightly against the rod by means of 
garter springs. The rings are held in 
accurately finished chambers, form- 
ing a casing around the piston rod. 

When packings are renewed, or 
when a new compressor is started, 
it is important to tighten regularly 
and evenly all round, as soft packing 
becomes softer through use. If the 
packings are screwed up tight at 
once, the rod will probably heat, the 
packing may be destroyed, and the 
rod will get scored. 

Oil Separator. The greater por- 
tion of the oil reaching the com- 
pressor cylinder passes out of the 
compressor with the refrigerant and 
must be separated out by means of 
an oil separator, for reasons given FIG. 165. Combined oil separa- 

i . tor and heater. 

later on. 

Figure 165 illustrates a typical oil separator, located in the 
engine room near the compressor. The hot gases enter through 
the tube (1) and leave the separation chamber (2) through the 
pipe (3). The oil is separated and accumulates in the bottom of 
the chamber (2), from which, by opening the cocks (4) and a 
needle valve (5), it is allowed to pass through the sight-feed 
arrangement (6) into the bottom of the chamber (7). Below this 
chamber is a heating chamber (8) through which hot water is 
passed, entering through pipe (9) and leaving through pipe (10). 
When the oil has been drained into the chamber (7), the needle 
valve (5) and shutoff cocks (4) are closed; the hot-water service is 





put on; and the heat frees the oil from the ammonia vapors, which 
are passed out through the pipe (11), leading into the suction pipe 
of the compressor. Having been freed from ammonia vapor, 
the oil is blown out through the drawoff cock (12) and pipe (13). 

In some oil separators a mechanically operated rotating plug 
is continuously transferring the separated oil from the separator 
(2) into the chamber (7). 

Great care should be taken in filtering and purifying oil 
reclaimed from the oil separator, as, if it is not entirely freed from 
impurities, the result when using it over again will be the wearing 
of the piston rod, piston, and cylinder walls. 

Expansion Valve. The expansion valve is fitted for the pur- 
pose of wiredrawing the refrigerant from the high pressure exist- 
ing before to the low pressure existing after the valve. It must 
therefore be capable of very fine adjustment. 

If any water gets mixed with the refrigerant, this water usually 
freezes in the expansion valve and clogs it; impurities have the 
same effect; for this reason it is very important that the expansion 
valve be kept absolutely clean. 

Dirt Catcher. In the suction line of the compressor is fitted a 
short piece of pipe provided with a sieve, for the purpose of 
preventing impurities, such as iron scale, little pieces of packing 
material, or even ice (frozen water) from entering the compressor. 
This sieve should be examined every day or so in the case of a 
new compressor installation, every 3 days for the next couple of 
weeks, and later on once every 2 months. If the sieve in the 
dirt catcher is allowed to get full of impurities, it may unexpect- 
edly burst. All the impurities will at once be drawn into the 
compressor and almost certainly cause serious damage. 


The lubrication of the compressor piston and cylinder is 
brought about entirely by the oil carried through to the interior 
of the cylinder from the stuffing box by the piston rod. 

There are three principal methods of lubrication, viz.: 

Bath oiling system. 
Mechanically operated lubricator. 
Splash oiling. 

Bath Oiling System (Fig. 166). Oil is pumped continuously by 
means of a small oil pump through the pipe (1) into the gland (2) 



surrounding the piston rod (3) ; it overflows from the top through 
the pipe (4) back again into the oil container, reentering the oil 
pump and circulating afresh. 

Figures 167 and 168 show a bath oiling system, where the oil is 
not circulated but seals the gland by maintaining a height of oil 
above the lantern in the gland. In Fig. 168 the oil flows to the 
gland under the full compression pressure. 

FIG. 166. Bath oiling system for stuffing box. 

When the stuffing-box gland packing is too loosely adjusted, an 
excess amount of oil will find its way into the compressor. On 
the other hand, if the packing is adjusted too tightly, too little 
oil will reach the interior; the piston rod will heat and may even 
become scored through the extreme friction and pressure exerted 
upon the rod in the gland ; also, the packing will suffer f r^m the 
heat; cotton or leather becomes brittle, and small portions may be 
carried into the cylinder, causing excessive wear. 



Mechanically Operated Lubricator. In dry-compression ma- 
chines experience has proved the necessity of employing mechani- 
cally operated force-feed lubricators, which will introduce a small 
quantity of oil with precision and regularity and in which the 

1 Hand Pomp 
3 Gland 
3 Oil Tank 

FIG. 167. Bath oiling 
system with overhead 

Fia. 168. High-pressure bath oiling system. 

oil feed can be adjusted to a nicety. The mechanically operated 
lubricator is driven from a moving part of the engine, therefore 
starts and stops feeding with the engine and delivers the required 
quantity of oil under pressure into the interior of the stuffing box. 
Whereas the lubricating oil in wet-compression machines readily 
adheres to the moderately warm piston rod and thus ensures 
sufficient oil's reaching the interior, the case is different in dry- 
compression machines. Because of the high temperature of the 
piston rod, the oil film will be thin, and very little oil will reach 



the interior of the compressor, unless it is introduced into the 
packing under pressure. That is the reason why mechanically 
operated force-feed lubrioators are used, as in this way the oil is 
certain to be carried along the piston rod into the compressor, and 

1 Hand Pump 

2 Oil Chamber 

3 Pipe to Gland 


FIG. 169. CO 2 pressure lubricator. 

the oil feed can be adjusted to the correct amount required for 
piston lubrication. 

It is important that a type of mechanically operated lubricator 
be used that will feed oil only. Several types of mechanically 
operated -lubricators feed air with the oil; the air thus introduced 
into the stuffing box is drawn into the compressor, increasing 
the pressure in the whole plant and considerably decreasing the 


Figure 169 illustrates diagrammatically a lubricator for CO 2 
machines which delivers the oil into the gland under pressure, 
the same as a mechanically operated lubricator, but is not capable 
of such accurate adjustment or control. It consists of a cylinder 
in which there is a piston with a piston rod. The one side of the 
piston is subjected to the condenser pressure of, say, 70 atmos- 
pheres. The other side (where the piston rod is) holds the oil 
and has an outlet fitted with an adjustable throttle valve, through 
which the oil passes out to the gland. 

The difference in area between the two sides of the piston is 
about 10 per cent and causes the overpressure which forces 
the oil into the gland. By this method leakage of CO 2 from the 
gland is entirely obviated, and the outer gland is only required to 
prevent leakage of lubricant. 

Splash Oiling. Some few makes of small vertical enclosed-type 
ammonia and CO2 compressors have a bath of oil in the crank- 
case, into which the crankpin bearing dips and splashes the oil 
to all parts. 

The oil level should be a little below the underside of the 
crankshaft; if it is too high, the oil will froth with the vapors of 
the refrigerant, which are continuously drawn through the crank 


The objects of internal lubrication of a refrigerating compressor 
are (1) to lubricate; (2) to form an oil-sealing film so as to prevent 
leakage of refrigerant past the piston and out through the stuffing 
box; (3) to preserve the leather or rubber which may form part 
of the packing material. 

If too little oil is used, the oil film will not be complete, so that 
excessive friction and wear take place, and leakage past the 
piston and piston rod occurs. 

If too much oil reaches the interior of the compressor, or if the 
separator is not sufficiently effective, a fair amount of oil will be 
carried over into the condenser and through the expansion valve 
into the evaporator coils, where, owing to the low temperature, 
it becomes sluggish and is only slowly carried through the coils 
back again to the compressor. 

The 0t7, in passing through the compressor, is exposed to the 
effect of the ammonia or carbonic acid, under moderately high 


temperatures. The result is more or less decomposition indi- 
cated by a darkening in color and an increase in gravity and 

Where the bath oiling system is employed, the system must 
be recharged with fresh oil, say once every year. 

Oils with too high a cold test, when carried over into the separa- 
tor coils, congeal on the inside of the tubes, and, as oil is a bad 
heat conductor, the capacity of the condenser and evaporator will 
be appreciably reduced. It is therefore important that the oil 
should possess a sufficiently low cold test so as not to become too 
sluggish to flow if it is carried over into the evaporator coils. 

The oil must not contain any moisture, as the moisture will 
freeze and cause it to congeal. For this reason, engine attendants 
should be warned not to put the oilcans near suction pipes covered 
with snow in the vicinity of the compressor, as water dropping 
from the outside of such pipes may drop into the cans and con- 
taminate the oil. Care should also be taken that the save-alls 
fitted under compressor glands and elsewhere to receive the waste 
oil are so made that no water dropping from the outside of the 
suction pipes can get into them and mix with the oil. Cases 
have been known where so much congealed oil has accumulated 
in the coils of the evaporator that they have become almost 

The oil for splash-oiled vertical compressors must have a very 
low viscosity, as otherwise it froths with the refrigerant; the 
froth fills the crank chamber, passes the piston, and clogs the 
whole system. 

Even under the best conditions a slight amount of oil will 
certainly get over into the evaporator coils in all types of com- 
pressors, and it is therefore advisable to have the coils thoroughly 
and regularly cleaned. They are best cleaned by blowing through 
dry steam and afterward air to dry the pipes. In bad cases, this 
treatment may be preceded by the application of a solution of 
soda ash. 

Glycerin is used as a lubricant where the packing of the 
piston or stuffing box consists partly of rubber, which in time 
is destroyed by mineral oil, whereas glycerin has no appreciable 
deleterious effect on rubber or leather. With a packing contain- 
ing only leather and brass or white metal, as in Figs. 163 and 164, 
mineral lubricating oils are used and, if properly selected and of 


good quality, will be found superior to glycerin in reducing the 
piston and gland friction. 

With efficient lubrication the piston rod assumes a smooth 
glossy surface, covered with a thin clean film of oil, and the 
piston rod maintains a moderate temperature. The stuffing 
box, as well as the piston, will be perfectly sealed, so that no 
leakage of refrigerant occurs. 


Where irregularities occur in a refrigerating plant, the effect 
is always to reduce its capacity for producing cold. The cause 
of the irregularity is not always easy to trace. 

The following are typical causes of trouble: 

Broken valve springs, preventing the valves from operating. 

Leaky valves. 

Leaky piston, piston rings out of order. 

Dirty condenser coils: deposit from dirty cooling water on the 
outside, or a coating of oil on the inside of the tubes. 

Expansion valve clogged or frozen (due to moisture in the oil, 
the use of poor cold-test oil, or impurities from the pipes). 

Inefficient operation of the evaporator (due to oil or moisture's 
congealing on the inside of the tubes or to salt incrustations 
from the brine on the outside of the tubes). 

Too little ammonia or carbonic acid in the system (due to leak- 
age from the pipes or stuffing box). 

The presence of air in the system, usually indicated by too 
high pressure in the system (air admitted through stuffing box). 


A number of ice-making plants ashore are steam driven. The 
steam after passing through the steam engine is condensed and 
subsequently used for the manufacture of can ice. Plate ice is 
not made from condensed steam. 

The cylinder oil used for internal lubrication of the steam 
engine may find its way into the ice, and this is most objection- 
able, as it discolors the ice and gives it an unpleasant odor and 

Troubles with discoloration of the ice can, however, be traced to 
a variety of causes, such as the boiler water, the raw "make-up" 
water, the exhaust-steam oil separator, and the water filters. 


Water. The water used as boiler feed water for the boiler 
must be specially selected; it must not be hard; it must be free 
from sodium, calcium, or magnesium; it should be neither acid 
nor alkaline or, at the most, should show only a slight 

The boiler should be of ample capacity for developing the 
amount of steam required; the water level in it should never be 
allowed to rise too high and should be as constant as possible; 
otherwise, priming of the boiler occurs, and salts in solution and 
impurities will be carried over with the steam into the steam 
engine and finally mix with the water used for ice making. 

Unsuitable water, used in the boiler or as raw make-up water, 
produces discolored ice. 

Oil Separator. It is the duty of the oil separator to remove as 
much as possible of the oil contained in the exhaust steam. The 
oil and moisture from the steam separate out in the bottom of the 
separator and are removed at regular and frequent intervals, 
automatically or nonautomatically. Most of the oil not caught 
by the oil separator will separate out in the reboiler (in which the 
water is heated and freed from air bubbles) and is automatically 
skimmed off the surface. Any traces of oil still left should be 
caught in the water filters (coke, charcoal, or gravel). 

Filters. Rust from the pipes and impurities of various kinds 
gradually accumulate in the filters. It is therefore neces- 
sary to clean or renew the filtering material at least once every 

Steam-cylinder Oil. When using compounded steam-cylinder 
oils, particularly dark cylinder oils, some of the oil may pass 
not only the oil separator but also the reboiler and even the 
filters, finally appearing in the ice. 

Filtered cylinder oils separate easily from the water. A good 
grade of filtered cylinder oil is therefore to be recommended for 
steam engines in ice-making plants; and if a good type of mechani- 
cally operated lubricator is employed, so that the oil can be 
introduced in the best manner and used economically, the use of 
a compounded filtered cylinder oil in place of a straight mineral 
oil is permissible, containing not more than 3 per cent acidless tal- 
low oil, as the admixture of tallow oil will enable the cylinder 
oil to be used exceedingly economically, and better lubrication 
will be obtained. 



The little oil that will be present in the exhaust steam is easily 
taken care of by the oil separator or, at any rate, by the reboiler 
or filters. 


Refrigerator oils must be straight mineral oils. Compounded 
oils have too high a setting point; they combine to some extent 
with ammonia, more or less saponification taking place, and they 
are rather inclined to absorb moisture from the atmosphere, 
which is very undesirable. Only low viscosity is required, the 
setting point being of supreme importance. 

For most refrigerators in ice-making plants an oil with a zero 
cold test will be satisfactory, but many CO 2 machines operate 
with lower evaporator temperatures than is the case in ice- 
making plants. A cold test of 25 F. will, however, satisfy the 
vast bulk of refrigerating compressors. Many ice-making plants 
prefer to use an oil with a better cold test than 0F. in order 
to have an extra margin of safety against the oil's congealing in 
the system. 

Oils specially low in viscosity are required for small splash-oiled 
vertical compressors to avoid frothing. 

To withstand the heat in the compressor without serious decom- 
position, refrigerator oils should be highly refined and highly 
filtered, pale-colored, straight-run distillates, containing as few 
unsaturated hydrocarbons as possible. 

Experience proves that three mineral refrigerator oils having 
the viscosities, setting points, and open flash points shown below 
will satisfy all requirements. 



at 50C. in 

Setting point, 

Flash point open, 


4 to 6 
1 to2 

13 to 20 
13 to 20 
4.5 to 8 

Below -40 

340 to 380 
320 to 360 
Above 300 

* See table, p. 57. 

These three mineral oils and glycerin are recommended for 
the following types of compressors: 



For Refrigerating Compressors 


For compressor 

For bearings 

Cold air 


Bearing oilf 4 

Ammonia,* open type 

Refrigerator oil 1 

Bearing oil f 3 or 4 

Carbonic acid, * open type 

or 2 
Refrigerator oil 2 

Bearing oilf 3 or 4 

Small, vertical, enclosed type, splash 
oiled, whether ammonia or carbonic 

Refrigerator oil 3 

Refrigerator oil 3 

Large, vertical, enclosed type, splash 
oiled, whether ammonia or carbonic 

Refrigerator oil 2 

Refrigerator oil 2 

Methyl chloride and. similar small 

Refrigerator oil 2 

Refrigerator oil 2 

* When rubber forms part of the packing in the stuffing-box gland, glycerin must be used 
and not mineral refrigerator oil. 
t For bearing oils see page 135. 



In order that the reader may understand and appreciate the 
lubricating problems met with in connection with gas engines, 
it becomes necessary to give first of all a picture of the mechanical 
and operating conditions. 

The subject has been divided into " Small and Medium-size 
Horizontal Gas Engines/' "Large Horizontal Gas Engines/' 
and "Vertical Gas Engines." For each group of engines particu- 
lars of typical engines and the methods of lubrication are given. 
Some information follows in regard to cooling and gas, both of 
which have an important bearing upon the lubrication of gas 
engines. The formation of carbon deposits is then treated in 
detail, and finally the important part played by the oil itself, 
and the correct grades to be recommended for the principal types 
of engines. 


Classification. Small horizontal gas engines have only one 
cylinder; they develop from 1 to 50 hp. and operate at high 
speeds, ranging from 500 to 190 r.p.m. Medium-size horizontal 
gas engines are made with one or two cylinders; the power 
developed per cylinder ranges from 50 to 250 hp., and the speeds 
range from 190 to 130 r.p.m. 

The cylinders are always water cooled, and where the power 
per cylinder exceeds 150 hp. the piston must also be water cooled. 
Indeed, many builders employ water-cooled pistons in engines 
ranging in sizes from 80 hp. per cylinder upward. 

Small and medium-size engines are practically always of the 
four-stroke-cycle type and, in view of the foregoing, may be classi- 
fied as shown in the table on page 463. 

The cylinders of medium-size engines may be arranged in 
various ways (Fig. 170) : 

Two cylinders, opposed the opposed-type engine (Fig. 170A). 




Two cylinders, one behind the other the tandem engine 
(Fig. 170). 

Two cylinders, side by side the twin engine. 


Number of cylinders 



Small size .... 


1 to 50 

500 to 190 

Medium size (without water- 
cooled pistons) 

Usually Ij sometimes 

50 to 150 

190 to 140 

Medium size (with water-cooled 

2, 3, or 4 
Usually 1; sometimes 

80 to 250 

180 to 130 

2, 3, or 4 

Three or four cylinders, side by side the multiple-cylinder 
engine (Fig. 170C). 

FIG. 170. Types of horizontal gas engines. 

Principle of Operation. The four-stroke-cycle principle of operation may 
be described as follows: 

First or Suction Stroke. Gas and air, constituting the fuel charge, are 
sucked into the cylinder through the open inlet valve as the piston moves 
away from the cylinder head. The exhaust valve is closed. 

Second or Compression Stroke. The piston moving toward the cylinder 
head compresses the fuel charge. Both inlet and exhaust valves are closed. 

Third or Power Stroke. Ignition by the spark of the compressed fuel 
charge produces explosion and expansion of the gases, forcing tf.e piston 
away from the cylinder head during the power stroke. Both inlet and exhaust 
valves are closed. 


Fourth or Exhaust Stroke. The piston moving toward the cylinder head 
drives the burned gases out through the open exhaust valve. The inlet 
valve is closed. 

Thus the four strokes of the piston, i.e., one power and three preparatory 
strokes, complete the cycle of events; hence the expression four-stroke cycle. 

Methods of Lubrication. Main Bearings. These are generally 
ring-oiled bearings, having an oil reservoir from which one or two 
revolving oil rings continuously carry the oil to the bearing sur- 
faces. In some medium-size gas engines the main bearings are, 
however, fed by gravity from an elevated tank and kept in con- 
tinuous circulation by a pump. 

Crankpin Bearing. This is lubricated by means of the well- 
known banjo arrangement. Either the oil is fed into the banjo 
from a sight-feed drop oiler, or the feed may come from a mechani- 
cally operated lubricator. 

Piston. Small engines are often fitted with only one oiler to 
supply the piston and wrist pin. The surplus oil on the piston 
collects in a groove at the top and through a tube drops into the 
wrist-pin bearing, more or less contaminated with carbon. With 
this method it is always necessary to overfeed the piston in order 
to ensure a reasonable supply of oil's reaching the wrist pin. 
Many small engines and most medium-size engines have therefore 
separate oilers for piston and the wrist pin, so that the right 
amount of oil can be distributed. 

The piston oiler, if it be a sight-feed drop oiler, should prefer- 
ably be provided with a ball check valve .to prevent blowback of 
escaping gases into the sight feed (see Fig. 18, page 116). 

Gravity sight-feed drop oilers will vary from, say, 40 drops per 
minute when nearly full to 16 drops per minute when nearly 
empty. If the quantity fed at a lower level is sufficient, as it 
must be if the engine is not to suffer, the extra quantity fed when 
the container is full is sheer waste and, in the case of the cylinder, 
positively detrimental. The oil feed is very susceptible to tem- 
perature changes, and the needle valves easily choke with dirt. 
The idea has therefore been steadily gaining ground that some- 
thing more reliable is needed, and the foremost engine builders 
are adopting a centrally placed multifeed lubricator operated 
mechanically from the camshaft. The lubricator should be 
designed on principles that will ensure a constant uniform oil 
feed, independent of the height of the oil level in the container 



and independent of the viscosity of the oil. Each feed should 
be from a separate pump unit, independent of the other feeds, and 
should preferably have a sight-feed arrangement showing the oil 
on its way to the engine; also, the feeds should be capable of being 
flushed and of instantaneous adjustment between wide limits. 

The lubricator usually has an operating lever actuated by a 
cam on the camshaft; the oscillating lever gives motion to the 
internal mechanism in the lubricator, so that pump plungers 
automatically pump oil through the various oil pipes. The 
individual feeds of the lubricator, once adjusted, require no fur- 
ther attention. 

In order to ensure that the oil pipes shall be always full, they 
should be provided at the extreme ends with check valves. 

FIG. 171. Timed oil injection to piston of gas engine by a mechanical lubricator. 

When the lubricator is stopped, and the pumps cease to operate, 
the check valves prevent the oil from running out of the pipes. 
The pipes are thus kept constantly full of oil, and instant lubrica- 
tion is assured whenever the engine and, consequently, the 
lubricator starts to operate. 

Timed Oil Injection to the Piston (Fig. 171). When the 
mechanical lubricator is designed to pump oil only and not oil 
and air, the piston oil feed can be timed to inject the oil under 
pressure at the right moment and to the ideal place, which is 
between the first and second piston ring, when the piston is at its 
most outward position. This enables the piston to carry the oil 
well into the cylinder. Feeding the oil in this way, the piston 
will act as an oil distributor; cleaner and more economical fabrica- 
tion of the piston is obtained, and practically no oil runs io waste 
from the lip of the liner. It is necessary that the oil should be 



1 Water Space 

2 Check Valve 

3 Oil Injector 

fed through a combined check valve and oil injector (see Fig. 172), 
the end of which barely touches the piston, so that for every 
impulse of the oil pump a small portion of oil is wiped off by 
the piston and taken right into the inner portion of the cylinder, 
distributing itself uniformly over the entire surface. Any devia- 
tion from this practice, either by feeding the oil nearer the front 
of the liner than indicated or by feeding it through a lubricator 
that does not time the injection of the oil, will mean a larger 

oil consumption (waste), more 
carbon deposit, and a lower mar- 
gin of safety. 

On the Continent, practically 
all gas engines are fitted with 
some kind of mechanical lubri- 
cator, so as to ensure that the oil 
feed to the piston shall be as 
regular as possible, but the 
importance of timed injection 
does not appear to be fully real- 
ized. The installation of such a 
mechanical lubricator means 
extra initial cost to the consumer 
and to the engine builder, but it 
also means a saving in oil con- 
sumption as compared with 
sight-feed drop oilers of as much 
as 50 per cent, and, what is much 
more important, it means a 
greatly increased margin of safety, as well as cleaner and more 
efficient lubrication throughout. 

One engine builder who had for years been using sight-feed 
drop oilers in connection with pumps (the oil dropping from the 
sight-feed drop oilers into the oil pumps) found that after install- 
ing mechanical lubricators of a good make, practically all their 
trouble during the guarantee period of their engines ceased. 
In many cases, engine attendants forget to adjust the sight-feed 
drop oilers, forget to start them or stop them, or allow them 
to run empty. With a mechanically operated lubricator there 
is only one container to fill, and one filling of the container 
will last several days; less attention is therefore required! 

FIG. 172. Oil injector. 



Valve Spindles and Cams. The valve spindles of inlet and 
exhaust valves (as well as the cams) are sparingly hand oiled, 
but in the case of larger gas engines, say above 50 hp., it is 
becoming general practice to lubricate the exhaust-valve spindle 
by one of the feeds from the mechanically operated lubricator. 
The feed must be very sparing and absolutely uniform. With an 
excessive oil feed the excess oil burns and carbonizes. With 
too sparing a supply of oil, too little lubrication is provided. 
The spindle becomes overheated and carbonizes what little oil 
it gets. In either case the exhaust-valve spindle will be inclined 
to " stick. 7 ' 


Large gas engines are used for driving electric generators in 
iron- and steelworks, in collieries, occasionally in large central 
power stations, and, in rare instances, in textile mills. 

Large two-stroke cycle gas engines are also extensively used 
in ironworks to drive blowing engines which produce compressed 
air for the blast furnaces. 

All large gas engines are double acting; most of them are of 
the four-stroke cycle type, built usually as tandem-cylinder 
engines, rarely with one cylinder only. The largest power units 
consist of two tandem engines placed side by side a twin-tandem 
engine operating an electric generator mounted between them 
on the main shaft. Two-stroke cycle gas engines have only one 
cylinder and operate at a lower speed than four-stroke cycle 



of cylinders 

per cylinder 


Four-stroke cycle double acting 

1 to 4 

300 to 1 500 


Two-stroke cycle double acting 


400 to 2,000 

10 %0 


Internal Lubrication. The cylinder, stuffing boxes and 
exhaust valve spindles are always lubricated by mejftns of a 
mechanically operated lubricator delivering the oil under pressure 
to the various parts and operated from the camshaft. 



Cylinder. The oil is introduced into the cylinder at three to 
six points, through ^-in. copper pipes from the mechanically 
operated lubricator. The oil inlets are sometimes located at the 
middle of the cylinder of four-stroke cycle engines, but in the 
case of two-stroke cycle engines this is not possible on account 
of the exhaust belt around the middle of the cylinder. In this 
case the oil inlets are placed about halfway between the exhaust 
belt and the cylinder ends. 

It is important that the oil be introduced into the cylinder 
at the moment when it will be fed directly to the piston and the 

FIG. 173. Stuffing box for large gas engine. 

piston rings. If introduced when the oil inlets are uncovered, 
it is burned by the hot gases, resulting in waste of oil and the 
formation of deposits. 

Stuffing Boxes. The stuffing boxes located in the cylinder 
heads are the parts that usually give the most trouble. The 
oil is introduced under pressure into the middle of the stuffing 
box and distributed over the entire frictional surfaces of the 
packing rings (see Fig. 173). These are usually cast-iron rings 
made in two halves and held together around the piston rod by 
means of light garter springs. Outside the packing rings there 
is occasionally a second stuffing box, employing soft packing. 

Exhaust-valve Spindles. Although the exhaust-valve guides 
surrounding the exhaust-valve spindles are water cooled, it 


becomes necessary to lubricate the spindles with a uniform and 
very sparing supply of oil for the same reasons as mentioned 
under medium-size gas engines. 

Mixing and Inlet Valves. The mixing- and inlet-valve spindles 
are sparingly hand oiled, except in very large engines, where they 
are supplied with oil through separate feeds from the mechani- 
cally operated lubricator. 

Gas and Air Pumps. As the gas and air are sucked into these 
pumps at a slight vacuum and delivered from the pumps to the 
working cylinders at a pressure of 4 to 6 lb., the temperature of 
the pump-cylinder walls, owing to compression, is not much 
above 100F. There is, therefore, no need for water-jacketing 
these cylinders, and their lubrication presents no difficulties where 
the gas and air are clean. 

The practice has been to feed the oil through sight-feed drop 
oilers into the center of each pump cylinder, with additional 
oil feeds to either end of both gas- and air- valve chambers. Fre- 
quently, however, an accumulation of deposits has been experi- 
enced due to moist dirty gas and overfeeding of the oil, the 
impurities adhering to the excess oil. Under these conditions it 
has been found better practice not to lubricate the pump cylinder 
and valve direct but to introduce the oil uniformly and sparingly, 
by means of a mechanically operated lubricator, through atom- 
izers in the respective intake pipes. 

External Lubrication. Circulation System. In the external 
lubrication of large gas engines a circulation system is usually 
employed. The lubrication of main bearings, crankpin, cross- 
head, tail-rod support, and guides is accomplished by means of 
oil fed by gravity from a top supply tank through a distributing 
pipe and its branches, into the various bearings. Adjusting 
valves are fitted in the branch pipes to regulate the oil feeds. 
Having done its work, the oil drains back to the bottom receiving 

An oil pump driven by the engine draws the oil from the receiv- 
ing tank and delivers it through an oil cooler into the top tank. 
If more oil is delivered to the top tank than is required for the 
bearings, the surplus oil overflows through an overflow pipe back 
into the receiving tank. The top tank may be omitted, and the 
oil passed directly from the oil cooler into the distributing pipes 
in which case it becomes necessary to have a relief valve, through 
which the surplus oil is passed back into the bottom tank. 


The oil is delivered to the crankpin through the hollow crank- 
shaft, and in order to distribute it well there are usually three or 
four radial holes (120 deg. apart) through which it reaches the 
large bearing surface. 

The oil for the crosshead bearings is delivered to the crosshead 
guide. The crosshead shoe is so long that the oil hole in the 
guide is never uncovered; consequently, the oil is enabled to 
force its way through drilled passages in the crosshead, as 
indicated in Fig. 174. 

To give an idea of the dimensions of bearings and oiling system, 
the following two examples are cited as typical of existing 
engines : 


Engine speed 94 r.p.m. 

Quantity of oil in circulation 40 gal. 

Type and size of pump Plunger pump, lj^ in. diameter 

by 1^2 in. stroke 

Speed of oil pump 94 strokes per minute 

Height of oil pump above oil level in bot- 
tom tank 4 ft. 

Three main bearings 18 in. diameter by 33 in. length 

One crankpin 18 in. diameter by 30 in. length 

Oil-delivery pipe 1^4 in. diameter 

Oil-return pipe 2 in. diameter 

All waste oil flows direct to the bottom oil tank, which has two 
vertical strainers, through which the oil passes to the oil pump; 
the suction pipe is fitted with a strainer and a nonreturn valve 
(foot valve). 


Engine speed 140 r.p.m. 

Quantity of oil in circulation 80 gal. 

Type of oil pump Rotary, 4 in. wheels 

Speed of oil pump 140 r.p.m. 

Height of pump above oil level in bot- 
tom tank 6 ft. 

Three main bearings 16 in. diameter by 25.5 in. length 

One crankpin 15.5 in. diameter by 19.0 in. length 

Two crosshead bearings for forked 

connecting rod 9 in. diameter by 9.5 in. length 

Oil-delivery pipe 2 in. diameter 

Oil-return pipe 2 in. diameter 



All waste oil is led to a separating tank, which retains most of 
the water and impurities. The oil then passes through a filter 
before being delivered to the bottom oil tank through a strainer. 

Timing shafts are usually supported by ring-oiled bearings. 

Eccentrics are equipped with sight-feed drop oilers or auto- 
matic-compression grease cups. 

FIG. 174. Crosshead of large gas engine. 

Valve levers are sparingly hand oiled. 

Governor. The governor is oiled partly by sight-feed drop 
oilers and partly by hand. 


Vertical gas engines are used principally for driving electric 
generators which produce current for lighting or for operating 
electric motors. They are also used, occasionally, for driving 
air compressors, large centrifugal pumps, refrigerating plants, 

Practically all vertical gas engines are of the four-stroke cycle 



Some large two-stroke cycle vertical gas engines have been 
developed in England but may be said to be still in the experi- 
mental stage. 



Number of 

power per 



Four-stroke cycle: 

Multiple cylinder 

1 to 6 

5 to 125 


Multiple tandem cylinder 

4 to 12 

5 to 125 

35% . 

Two-stroke cycle: 

Double acting, with oil or water-cooled 


2 to 4 

200 to 500 


Constructional Points. There are two types of vertical four- 
stroke cycle gas engines: the multiple cylinder type and the 
multiple tandem-cylinder type (shown diagram- 
matically, Fig. 175). Vertical engines are rarely 
built with one cylinder only; they generally have 
three or four. The multiple tandem-cylinder type 
usually has four or six pairs of cylinders, i.e., 8 or 
12 cylinders, and is developed only in England: 
by the British Westinghouse Company and by the 
National Gas Engine Company. 

As vertical gas engines are always enclosed, and 
the cylinder walls copiously supplied with oil, the 
pistons are frequently designed with oil scrapers 
at the bottom in connection with grooves from 
which the oil can be drained through holes to the 
interior of the piston and thence down into the 
crank chamber. Also, it is good practice to design 
the pistons in two parts, inserting between the top 
and bottom portion a, plate, which prevents the 
oil from the gudgeon pin from splashing into the 

hot hollow piston head, where it would otherwise 

Diagram of ver- burn and char. As an alternative the piston may 

tical tandem- | De cas ^ w ith a projecting internal lip above the 
type gas engine. . r " , . . j u , , 

gudgeon pin, the hole being closed by a plate. 
In the case of the vertical tandem-type gas engines, the two 
pistons are connected by means of a piston rod working in a 


sleeve between the two cylinders. The air below the top piston 
is compressed on the downstroke and acts like an air buffer. 

Methods of Lubrication. Vertical four-stroke cycle gas engines 
operate at high speed. In order to prevent the lubricating oil 
from splashing away from the bearings, all external-motion parts 
are enclosed in a crank chamber or casing, so that the parts may 
be copiously supplied with oil, either by the splash system of 
ubrication or by the force-feed circulation system. 


Splash Oiling System (Fig. 176). The lower portion of the 
crank chamber is filled with oil to the level indicated in the draw- 
ing, and an adjustable overflow pipe is fitted to one end of the 
crank chamber in order to maintain a correct oil level. The 
bottom ends of the connecting rods dip into the oil and splash it 
to all parts requiring lubrication. 

Oil is fed into the outer main bearings by sight-feed drop oilers. 
Leaving these bearings it drops into the crank chamber and 
thus makes up for the amount of oil that goes away as oil spray 
or is burned away inside the cylinders. In place of sight-feed 
drop oilers, a mechanically operated lubricator is preferably 
employed to supply oil uniformly to the main bearings, in this way 
making the lubrication system entirely self-contained and auto- 
matic in action. 

A correct constant oil level must be maintained, in order to 
secure uniform splash to all parts and to obtain greatest economy. 
Connecting rods should dip into the oil to the same depth, and 
the oil level should be lowered until the formation of carbon 
deposits on the pistons is reduced to a minimum. 

An irregular or too high oil level means waste of oil and excessive 
carbonization. Too low an oil level or the use of an oil too heavy 
in viscosity means unequal distribution and insufficient lubrication 
for some of the parts, resulting in excessive wear of those parts. 
It cannot be too strongly emphasized that an adjustable overflow 
should be installed, in order that the correct oil level, once 
established, may be automatically maintained. 

Force-feed Circulation System. This system delivers the 
oil under a pressure of from 5 to 20 Ib. per square inch to all 
bearings, the oil leaving the gudgeon pins splashing cu to the 
cylinder walls. The circulating oil sometimes passes a filter or a 



Oxidation. In passing through the main bearings, crankpin 
bearings, and wrist-pin bearings, the oil is subjected to high tem- 
perature and speed of the rubbing surfaces. Oxidation takes 

1 1 1 1 1 1 i i 


1 1 1 1 I 1 1 1 1 

1 1 1 1 1 1 1 i 


FIG. 176. Splash-oiled vertical gas engine. 

place and is indicated by a darkening in color, an increase in 
viscosity and gravity, and the development of acidity. 

Temperature. As the crank chamber is enclosed, the heat 
radiated from the pistons and cylinder walls is, to a large degree, 
retained, so that the oil in the crank chamber gets warm, reaching 
a temperature of from 100 to 160F. If a temperature of 140F. 
is greatly exceeded, the life of the oil will be shortened, and it may 


throw down a dark deposit caused by oxidation similar to that 
which may take place with turbine oil. 

The force-feed circulation system is always employed in verti- 
cal, four-stroke cycle, multiple tandem-cylinder gas engines and 
is superior to the splash lubricating system. The splash system 
is used in some multiple-cylinder engines, but the majority of 
these engines employ the force-feed circulation system. 

It is a common trouble that oil in the crank chamber becomes 
contaminated with carbonized matter working down from the 
pistons. In order to prevent dirty oil from the trunk pistons from 
dropping into the crank chamber, some builders of vertical two- 
stroke cycle gas engines and vertical multiple-cylinder four-stroke 
cycle engines raise the cylinders above the crank chamber by 
means of a distance piece. The piston rods pass through 
scraper glands and are connected inside the crank chamber to 

By this construction carbonized matter can be prevented from 
entering the crank chamber, but as the pistons are not lubricated 
by splash from the crank chamber, it becomes necessary to lubri- 
cate them independently, by means of a multiple-feed mechani- 
cally operated lubricator. This practice permits the use of a 
different oil for piston lubrication, which is often desirable. 

As burned gases occasionally escape past the pistons into the 
crank chamber, this is provided with an air-vent pipe, frequently 
fitted with a fan, which sucks away from the enclosed crank 
chamber fumes and vaporized oil. At the same time, cold air is 
constantly drawn through the engine and helps to cool the pistons 
and crank chamber. 

As regards the piston lubrication, one oil feed from the lubri- 
cator goes to each piston, delivering the oil through a check valve 
into an annular oil tube surrounding the cylinder, from which two, 
four, or six leads go through the water jacket and distribute the 
oil over the piston surface. As the oil is introduced at one point 
of the annular oil tube, it is likely to pass into the cylinder through 
the leads nearest this point, so that the opposite side of the cylin- 
der may get little or no oil direct from the leads. For this reason 
it is good practice to have each point of entrance to the cylinder 
fed by a separate oil pump, so that each feed can be controlled 
with certainty. Two oil feeds suffice up to 21-in. cylinders, one 
for the front and one for the back of each piston. 


Furthermore, the oil should preferably be introduced in line 
with the level between the first and second piston ring, when the 
piston is in its lowest position. The oil feeds from the mechani- 
cally operated lubricator should be capable of independent adjust- 
ment, so that the exact amount of oil required can be supplied to 
the sleeves and pistons. 

Pistons. The piston rings should be in good condition and 
pegged, so that only a sufficient amount of oil for full lubrication 
will reach the entire piston surfaces. Too much oil splashed 
from the crank chamber (too high oil level, too high oil pressure) 
or too much fed directly to the piston means that excess oil 
will pass to the piston tops, where it will burn and char and ulti- 
mately form deposits. 

In the case of multiple tandem-cylinder engines, care should be 
taken that the quantity of oil fed from the mechanically operated 
lubricator to the top pistons and the sleeves be reduced to the 
exact amount required. This applies also to multiple-cylinder 
engines with a distance piece between the cylinders and the crank 


Cooling of all parts of the engine that come in contact with the 
hot gases is necessary. Without adequate cooling, unequal 
expansion and distortion of the overheated parts, excessive 
wear, and piston seizure would occur. 

Small gas engines employ the thermosiphon system, but most 
medium-size and all large gas engines employ pump circulation. 
In large gas engines, provision is made for adjusting the supply 
of cooling water to the various parts (cylinder walls, piston, 
piston rods, cylinder covers, and exhaust valves). All return- 
water pipes are carried to a central place, where the temperature 
and the quantity of cooling water from each part can be con- 
trolled. The cooling water, after its return from the various 
parts, is pumped through a cooling tower, cooled, and again 
pumped through the engine. 

Smaller engines employ cooling tanks, which must be so 
arranged that the water from one tank flows into the bottom of 
the next one (Fig. 177A). With the arrangement shown in 
Fig. 177 B the water flow is short-circuited, and the water is not 
properly cooled. 



In the inlet and outlet pipes should be fitted thermometers for 
registering the temperature of the cooling water. The average 
temperature of the return cooling water should be between 100 
and 130F. in large gas engines and between 100 and 140F. 
for smaller ones. If the outlet water from the water jacket 
is too high, the temperature of the cylinder wall will rise; the 
oil film thins out, losing its sealing power; and the explosion gases 
blow past the piston. If the outlet water is much below 100F., 
the cooling of the cylinder wall is too efficient. The oil film 
becomes sluggish, the oil spreads with difficulty, and a great deal 
of power is lost in overcoming the oil drag on the piston. A tem- 

FIG. 177. Cooling tanks. 


perature of 115 to 120F. is therefore preferable in order to 
ensure good piston seal and a free sliding motion of the piston. 

The cooling water must be clean, for, if impurities settle in the 
water jacket, the cooling of the cylinder walls and pistons (where 
pistons are water cooled) becomes defective; the temperature 
rises ; and preignition, caused by incandescent deposits inside the 
combustion chamber, is likely to occur. 

Where the gas contains an excessive amount of sulphur (large 
gas engines), successful results have been obtained by allowing 
the cooling water to run through the engine at a higher tempera- 
ture as high as 160F. The higher temperature of the cooling 
water minimizes, or entirely prevents, the condensation of mois- 
ture from the expanding gases and thereby the formation of 
sulphurous acid, which would attack the internal surfaces of the 
engine that come in contact with the gases. 


When using rich, highly inflammable gas, such as natural 
gas, the compression pressure at the end of the compression 



stroke must be proportionately low; otherwise, preignition will 
occur, due to the heat developed by compression. With weak, 
less inflammable gas, such as the producer gases, the compression 
pressure can be made much higher. 

This is shown in the table below, which gives average compara- 
tive heat values of gases and corresponding average compression 

Kind of gas 

Heat value, 

pounds per 
square inch 

Natural gas 



Illuminating gas 



Coke-oven gas 



Producer gases 



Blast-furnace gas 



Small and medium-size gas engines are operated by natural 
gas; illuminating, or town gas; suction producer gas; or pressure 
producer gas. Large gas engines are operated by blast-furnace 
gas, coke-oven gas, or pressure producer gas. 

Natural Gas. Natural gas is found in the oil districts of the 
United States, Canada, Russia, and Mexico. It is dry in its 
natural state, with a degree of purity that makes cleaning 

Illuminating, or Town, Gas. Illuminating gas is used practi- 
cally only for small gas engines. It is made from bituminous coal 
by dry distillation. It is free from impurities and is, therefore, 
an excellent fuel. 

Suction Producer Gas. Suction producer gas is usually made 
from coke, anthracite coal, lignite, wood refuse, etc. The engine 
draws the gas from the producer by suction hence the name 
suction producer gas. Suction producer gas plants are used for 
installations comprising one or more engines with a total power 
not exceeding 500 hp. 

Pressure Producer Gas. Pressure producer gas is made from 
a variety of fuels, such as bituminous coal, lignite, coke, anthra- 
cite, charcoal, sawdust, and wood refuse. It is produced under 
slight pressure hence the name pressure producer gas. 


Pressure producer-gas plants are sometimes used for installa- 
tions as small in size as 200 hp., but the installations usually 
range from 400 to 2,000 hp. or more, employing medium- or 
large-size gas engines. 

Where bituminous coal is used, rich in tarry matters, the clean- 
ing plant for the gas must be more elaborate and efficient and 
therefore more costly than in the case of suction producer-gas 
plants, which preferably use fuel free from tar. 

Producer gas, whether suction gas or pressure gas, should be 
thoroughly scrubbed, cooled, and cleaned; but, notwithstanding 
all precautions taken, the gas always contains, besides moisture, 
more or less impurities, such as soot, fine dust (coke dust, ash), 
and tar, which are carried into the engine arid interfere with 

Blast-furnace Gas. The blast-furnace gas coming from the 
blast furnace contains a great quantity of impurities, consisting 
df lime dust, fine iron oxide, coke dust and volatile matter from 
incomplete combustion in the blast furnace, water impurities 
from the water used in washing the gas, and a small amount of 
sulphur. The quantity of impurities varies from 12 to 25 g. 
per cubic meter of gas and is reduced in the cleaning plant to 0.01 
to 0.03 g. per cubic meter. If the impurities are more than 0.05 g. 
per cubic meter, the gas is dangerous for the engines and will 
cause deposits and excessive wear of piston rings and cylinder 

The gas is freed from the dust and tarry impurities by either 
the wet- or the dry-cleaning process. By the latter it is filtered 
dry through filter bags and is delivered in a less moist condition 
to the engines, so that it is less liable to deposit dust in the 
mixing valves and cylinders. 

Coke-oven Gas. Coke-oven gas is produced from bituminous 
coal during the production of coke. A portion of the gas is used 
in the coke oven, but the surplus can be used for operating gas 
engines. The coke-oven gas contains, besides coke dust, tar and 

Sulphur in Coke-oven or Producer Gas. Heavy wear is often 
noticed when the gas is not sufficiently low in sulphur content. 
The wear is chiefly on the piston rod but only on the part rubbing 
in contact with the rings in the metallic packing. The greatest 
wear is where the rod is coolest, i.e., where the water enters; the 


wear is less on the tail rod, where the water leaves the rod warm. 
The rod does not get pitted but wears uniformly, maintaining a 
bright polished surface, with dark-colored patches showing here 
and there. 

By allowing the cooling water to pass warmer through the 
rods, the wear, as mentioned on page 477, is reduced or eliminated. 

If a piston rod gets splashed with water from the water-dis- 
charge tubes (outside the cylinder), it will wear rapidly, if the gas 
contains an excessive amount of sulphur. A water leak from the 
cylinder head into the metallic packing will have a similar effect. 
In the case of a porous cylinder, allowing cooling water to leak 
into the cylinder, sulphur will cause extremely rapid wear. 


Deposits form in the mixing and inlet valves, in the cylinders, 
on the piston, behind and between the piston rings, on the exhaust 
valves, and in the stuffing boxes. In two-stroke cycle engines 
deposits are also formed in the gas and air pumps. Deposits may 
arise from one or several of the following causes: dust or dirt in 
the intake air, incomplete combustion, impurities in the gas, 
overfeeding of oil, the use of an unsuitable oil. The formation 
of deposits under certain conditions leads to preignition and 

Intake air is usually not filtered, even when the engines are 
placed in dirty surroundings. Impure intake air is therefore 
a frequent cause of deposits in gas engines, regardless of the kind 
of gas used (see Examples 33, 36, and 38, pages 483, 484, 486). 
In such cases a chemical examination will prove the presence or 
sand, brick dust, lime dust, etc. The deposit will also contain 
oil and partly decomposed oil, owing to the oxidizing action of 
the impurities on the oil under high-temperature conditions, and 
there will always be present a percentage of iron and iron oxide, 
due to wear of the piston rings and cylinder. 

In large gas engines the air should be filtered through coarse 
canvas or similar material before passing to a large settling 
chamber, which will collect more of the solid impurities. 

Incomplete combustion will bring about sooty crumbly deposits 
and may be due to poor ignition or improper timing of the 

Impurities in the Gas. (See Examples 36, 37, and 38.) 


Where producer gas is in use, deposits may be caused by such 
impurities as ash, fine coke dust, free soot, or tar passing into 
the engine. All fuels contain ash, and a regular feeding of fuel 
through the generator and removal of clinker from the grates 
are very desirable, because, if the grate is not covered with a suffi- 
cient layer of fuel, fine ash is likely to be carried over with the 
gas. Regular firing is therefore important, as it prevents the 
layer of fuel from getting too low. 

Where coke is used, there is no tar, but coke dust may be 
carried over in such fine form that the water trap, scrubber, and 
filter will not remove it. 

Where gas is produced from anthracite coal, tar and soot 
may both be carried over, although anthracite contains only a 
small percentage of tar. 

Where gas is produced from lignite, which contains more tar 
and soot, the danger of forming deposits inside the engine is 
more pronounced. 

Lignite also contains a small percentage of sulphur; this, in 
many cases, will cause a blackening of the piston surface but 
rarely causes serious trouble. 

Where gas is produced from bituminous coal, which contains 
a large percentage of tar, there is greatly increased likelihood of 
the gas's carrying soot and tar into the engine. 

Coke-oven gas and pressure producer gas contain some vola- 
tilized tar which cannot be eliminated in the producer plant 
and settles in the gas-inlet valve and mixing chamber or in the 
gas pump (two-stroke engines). When inlet valves stick, they 
can be "freed" by applying creosote. The tar affects the 
lubrication, encouraging the formation of carbonaceous deposits. 
It is this formation of tar that makes it necessary in suction 
producer plants to employ coke or noncaking anthracite coal. 

The moisture in wet gas, such as the producer gases and blast- 
furnace gas, forms a paste with the impurities in the air or gas, 
thus providing a base for the ready formation of deposits. The 
impurities collect in the mixing and inlet valves and on the 
internal surfaces of the engine exposed to the gas, adhering to and 
contaminating the oil film on the piston, piston rings, and cylinder 
walls. The pasty deposits in the mixing-valve chambers in 
time become crumbly, peel off, and are swept into the cylinders, 
causing excessive wear, preignition, etc. 


In two-stroke cycle gas engines, moist gas will also deposit 
the dust, in the form of a dark sludge, in the valve chamber of 
the gas pump. The sludge causes increased resistance in moving 
this valve, with consequent sluggish action of the governor. 

Deposits arising from air or gas, or both, always contain oil 
and also partly decomposed oil, the latter due to action of the 
impurities on the oil under high-temperature conditions. 

Overfeeding of Oil. The surplus oil, fed to the internal parts, 
burns and chars; it also attracts and collects the impurities from 
the gas and air, resulting in a dark-colored carbonaceous deposit 
of a harder or softer nature, depending upon the nature of the oil 
in use. Even with a good-quality oil in use, the oil feeds should 
be reduced to the exact amount required for full and efficient 
lubrication. This will lead to cleaner lubrication, as the impuri- 
ties find less oil to which they can adhere. 

Deposits accumulating behind the piston rings may cement 
the rings in their grooves, so that they lose their elasticity and 
break easily. Heavy wear takes place; the oil film is burned 
away ; the burning gases pass the piston, and, in the case of double- 
acting engines, pass from one side of the piston to the other, 
igniting the fuel charge on the opposite side and causing preigni- 
tion. Increased oil feed will only aggravate the trouble. Fre- 
quently, the stuffing boxes in large gas engines are overlubricated, 
with the result that carbon deposits are formed, causing the pack- 
ing rings to stick in their grooves. Wear follows, and the gases 
blow past the rings. 

Preignition. When carbon or other deposits develop inside 
the combustion chamber, and particularly if the water cooling 
is inefficient, the deposits often become incandescent and pre- 
ignitions occur, causing abnormally heavy strains on the engine. 

But deposits are not the only cause of preignition. Jointing 
material asbestos, red lead is often the cause of this trouble. 
Preignition may also be due to the use of rich gas, e.g., town gas 
in engines designed for suction gas, as the richer gas ignites 
spontaneously at a lower temperature. 

With blast-furnace gas it is difficult to prevent preignition, 
owing to the quantity of fine lime dust in the gas, which, when it 
settles inside the cylinders, easily becomes incandescent. 

In large engines preignitions occur every 2 to 3 hr. under the 
most favorable conditions and, under unfavorable conditions, 
every few revolutions. 



Preignitions may be caused by the explosion gases leaking past 
the piston rings from one side of the piston to the other. This 
happens when the rings are badly sealed, owing either to accumu- 
lated deposits causing the rings to be inflexible in their grooves, or 
to the use of too low viscosity lubricating oil, or to a furred up 
water jacket (the oil film gets hot and thins out), etc. 


Example 33. Dirty Intake Air. The following analysis of two 
deposits taken simultaneously from the piston indicates dirty 
intake air, which has caused a great deal of wear (iron oxide). 
The hard deposit at one time has no doubt passed through the 
stage represented by the soft deposit, the oil gradually charring 
and hardening the mass. The oil in use was a straight mineral 
paraffin-base oil; had it been an asphaltic-base oil, and preferably 
compounded, the deposit would not have hardened, but would 
have been in the form of a crumbly or greasy paste. 


per cent 

Components of deposit 






Volatile matter insoluble in petroleum spirits . . 
Iron oxide 

33. if 68 

54. 3 } 66 - 4 
30 4 




Balance undetermined 

2 4 




It is surprising many times to see the lack of care in not pro- 
viding gas engines with reasonably clean intake air. 

Example 34. A Curious Case of Spontaneous Ignition. An 
old gas engine of about 25 hp., with leaky piston rings and with 
a temperature of about 200F. in the water jacket, was using a 
heavily compounded oil. The gudgeon pin and piston were 
very hot. The engine was in a wooden shed close to saw benches, 
and, as- the door of the shed was always open, a considerable 
amount of wood dust was always entering the engine room. In 
particular where the crankpin had thrown the oil on the side of 
the shed, the dust and oil had formed a layer a Y in. thick. One 
day the heat of the piston caused the oil to ignite, throwing out a 



flame sufficiently long to reach the layer of oil on the side of the 
shed, which it also ignited. 

The ignition took place in the hollow part of the piston where 
the gudgeon pin end of the connecting rod works. The sawdust 
had accumulated in the hollow of the piston and was saturated 
with the highly compounded oil in use. As time went on, this oil 
became gummy, oxidizing more and more; and as the heat from 
the piston increased, owing to the gas engine's being heavily 
loaded, the oxidizing effect raised the temperature up to ignition 

Example 35. Bad Alignment. A 60-hp. gas engine was con- 
tinuallyb reaking piston rings; the back piston ring could not 
be lubricated, and the piston could therefore never be kept tight. 
The makers had overhauled the engine time and time again 
without locating the cause ; a special attachment was fitted which 
lubricated the back ring, but the breakages continued. Finally, 
the seat of the trouble was discovered; the cylinder was out of 
line with the crankpin to the extent of ^e i n -> which caused 
a great pressure between the piston rings and the liner. 

Example 36. Deposit Caused by Impurities in the Gas. 
Several large blast-furnace gas engines of the Cockerill type 
(double-acting, four-stroke, tandem engines, 1,500 hp.) were 
wearing badly owing to deposits which continued to develop 



Inside of 


Components of deposit 





the rings, 

the rings, 



Oil and moisture 





Volatile matter insoluble in pe- 

troleum spirit 


43 3 

37 8 

56 9 

Silicates and silica. . . ... 


12 1 

17 7 

12 5 

Iron oxide 




10 8 

Aluminum oxide 


6 4 

Calcium oxide 




10 2 

Balance, containing magnesium 

oxide and traces of other me- 

tallic salts 


3 2 







inside the cylinders. The cylinder wear had averaged 0.7 mm. 
per annum, as compared with the normal wear for large blast- 
furnace gas engines, which ranges from 0.3 to 0.5 mm. per annum. 

Samples of deposit were obtained from the points shown in 
the table on page 484 and analyzed. 

All the deposits were hard, granular, and black, and their com- 
position shows that the gas is mainly responsible for their forma- 
tion, the dust contents in the gas ranging from 0.025 to 0.035 g. 
per cubic meter. The lubricating oil in use was a deep-red 
straight mineral paraffin-base oil containing filtered cylinder 
stock which carbonized and cause the deposit to bake into hard 

1 Air Pit 

2 Filter Screen* 

3 Mr Intake Pipe 

FIG. 178. Filtering intake air. 

crusts. It is possible that some of the dust came with the intake 
air, as the intake-air pipes were placed with their ends turned up, 
so that the dirt and dust had free access. Besides, on this par- 
ticular side of the powerhouse were situated the stores of coal, 
coke, etc. It would have been better had the air pipes been 
placed on the opposite side of the powerhouse, taking the air 
through large settling chambers, fitted with suitable filter screens. 

Figure 178 shows a good arrangement for preventing impurities 
from entering a gas engine with the intake air. 

Example 37. Tarry Deposits. Two-stroke gas engines em- 
ploying coke-oven gas are often troubled with tarry deposits, 
which accumulate in the gas pumps, on the inlet valves, and 
inside the cylinders. 

A 500-hp. Koerting engine had to be dismantled every 4 weeks 
to clean away the deposits. A straight mineral dark-red paraffin- 


base oil was used, fed through sight-feed drop oilers. A mechani- 
cally operated lubricator was installed in order to bring about a 
regular feed, so that a minimum oil consumption could be estab- 
lished. At the same time an oil made chiefly from asphaltic-base 
mineral oil and compounded with 7 per cent of nut oil was intro- 
duced with most remarkable results, it being found possible to 
operate the engine for 5 months before the deposit had to be 
cleaned out. In addition, it was found that the engine now 
started quite easily on compressed air, whereas with the previous 
oil it was necessary to "motor" the generator when starting. 

Example 38. Deposit Due to Water Leakage. In the case of 
seven 800-hp. blast-furnace gas engines single-cylinder engines 
with open-ended trunk pistons a dark-brown, almost black, oily 
deposit was continuously working its way out from the cylinders. 
A sample showed the following analysis: 

Per Cent 

Oil 35.3 

Water 8.2 

Iron, iron oxide, and silica 21 .9 

Volatile matter insoluble in petroleum spirit 34.6 

The water came from the leaky cooling-pipe connections; the 
silica came in with the intake air or the gas and caused heavy 
wear, the action being accentuated in the presence of water. 


General. The frictional losses in a small or medium-size gas 
engine range from 15 to 30 per cent; in large gas engines, from 
10 to 20 per cent of the rated horsepower, this loss being constant 
and independent of the actual engine load. It is easy to waste 
from 5 to 10 per cent of the engine power in unnecessary friction 
by using unsuitable lubricating oil or an inefficient lubricating 


Piston Lubrication. The piston of the gas engine is the most 
vital part from a lubricating standpoint. With impure gas or 
unsuitable oil or overfeeding, deposits develop on the piston 
head and behind the piston rings and will appear in the form 


of a black, oily coating. These deposits soon cause the piston 
rings to cement in their grooves; the gases blow past the piston; 
excessive friction and wear take place on the piston rings and on 
the cylinder walls. 

It is important that the oil shall have a suitable viscosity. 
If an oil of too heavy viscosity is used, it does not spread easily 
over the piston surface. The friction is high owing to the heavy 
oil drag on the piston, and impurities in the gas or air will cling 
to the heavy oil and bake into crust-like deposits. 

If an oil of too light viscosity is used, it will break down under the 
influence of the high temperature in the cylinder; it will lose its 
sealing power, causing excessive friction and wear. Carbona- 
ceous deposits are formed which on account of the wear will be 
found to contain a large percentage of iron and iron oxides. 

Bearing Lubrication. Main bearings usually give no trouble; 
it is bad practice to add oil to ring-oiled main bearings daily. 
The bearings do not run cooler, but the oil is wasted, overflowing 
from the bearing ends. 

The bearing reservoirs should be emptied, cleaned, and 
recharged at regular intervals from 3 to 6 months depending 
upon the purity (absence of dust) of the air in the engine room. 

The crankpin is a very important part of the engine, as it 
transmits the power from the piston to the main shaft. Occa- 
sionally, a heavy-bodied oil is required for lubricating the crank- 
pin of medium-size engines owing to the heavy crankpin 
bearing pressures. As this oil may be too heavy in body for the 
cylinder lubrication, two different oils are sometimes used, 
although usually one is used throughout. 

The gudgeon or wrist pin requires particular care in lubrication, 
as it is located in the interior of the heated hollow trunk piston, 
where it is subjected to high temperature. The pressures on the 
wrist pin are high, and, as the oscillating motion of the connect- 
ing rod is slight, the oil spreads with difficulty over the bearing 
surfaces. Consequently, the lasting and lubricating properties 
of the oil used are very important. 

The blackened waste oil coming from the piston and wrist pin 
in small and medium-size engines should be arrested by a division 
plate (see Fig. 171, page 465) and drained away so that it will 
not run into the crank pit and contaminate the oil coming from 
the crankpin. 



Piston Lubrication. Owing to the large diameter of the pis- 
tons, it is of the greatest importance that the oil be introduced 
direct to them at several points and in such a manner that the 
correct amount is delivered at the correct moment, positively and 
regularly. Incorrect methods of lubrication, or lubricators that 
cannot be relied upon to feed the oil in the best manner, mean 
excessive oil consumption ; the waste of oil is less important than 
the fact that the impurities in the gas adhere to the surplus oil, 
which leads to the formation of carbonaceous deposits. 

In the United States practically all large gas engines are 
lubricated with oils that are highly filtered cylinder stock, having 
viscosities ranging from 26 to 32 centipoises at 100C. 

The very same types of engines are lubricated successfully 
in Europe with oils having viscosities ranging from 26 to 76 
centipoises at 50C. It is practically certain that the American 
engines would operate better, with less friction, easier starting, 
and less carbon deposit, if they used oils more in line with 
European practice. 

Stuffing-box Lubrication. When an oil too low in viscosity 
is used it cannot seal the packing properly and allows gas to 
blow through the stuffing boxes. The gas burns and chars the 
oil, causing heavy wear and the formation of deposits. 
The action of the gas is extremely erosive, cutting grooves in the 
piston rod and shortening the life of the metallic packing. The 
gas escaping from the stuffing boxes into the engine room is very 

It is quite usual to find that the oil is fed to the stuffing boxes 
by means of sight-feed drop oilers and that, therefore, a great 
deal more oil js used than when the stuffing boxes are lubricated 
by means of a mechanically operated lubricator. It is very 
important that the lubrication of the stuffing boxes should be 
kept clean and economical. Excessive oil feed means formation 
of carbon deposit, excessive friction, and wear, accentuated by 
continuous " blowing" of the glands. 

Where the stuffing boxes are worn, and blowing takes place, 
they should be put in good order at the earliest opportunity, 
although as a temporary arrangement an oil heavier in body may 
be used in order to seal the packing and prevent blowing. Satis- 


factory operation of the stuffing boxes is perhaps a more difficult 
problem from a lubricating point of view than that of any other 
part of the engine; it pays, therefore, to give special attention to 
selecting the correct oil for their lubrication and applying it in 
the best manner, using it as economically as possible. 

Only fresh oil should be used for lubrication of stuffing boxes, 
pistons, valves, etc. Any waste oil that may be collected from 
underneath the stuffing boxes, valve spindles, etc., may be treated 
in a heated separating tank and filter, after which it can be u